PhantomTroupe_Classification.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 152,
   "id": "3d964623",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np \n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.model_selection import train_test_split\n",
    "import sklearn.preprocessing as preprocessing \n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "import sklearn.metrics as metrics \n",
    "import sklearn.svm as svm \n",
    "import sklearn.naive_bayes as naive_bayes\n",
    "from sklearn.linear_model import LogisticRegression \n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.naive_bayes import GaussianNB, BernoulliNB\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.ensemble import AdaBoostClassifier\n",
    "import xgboost as xgb\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.ensemble import BaggingClassifier\n",
    "from sklearn.neural_network import MLPClassifier\n",
    "import seaborn as sns \n",
    "import json\n",
    "import nltk\n",
    "import string\n",
    "import re\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "0208e1c1",
   "metadata": {},
   "outputs": [],
   "source": [
    "main_df = pd.read_csv(\"train_data.csv\")\n",
    "test_df = pd.read_csv(\"test_data.csv\")\n",
    "\n",
    "main_df = main_df.append(test_df)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bab7c4c4",
   "metadata": {},
   "source": [
    "# 1. PreProcessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "aa1c721c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>url</th>\n",
       "      <th>webpageDescription</th>\n",
       "      <th>alchemy_category</th>\n",
       "      <th>alchemy_category_score</th>\n",
       "      <th>avgLinkWordLength</th>\n",
       "      <th>AvglinkWithOneCommonWord</th>\n",
       "      <th>AvglinkWithTwoCommonWord</th>\n",
       "      <th>AvglinkWithThreeCommonWord</th>\n",
       "      <th>AvglinkWithFourCommonWord</th>\n",
       "      <th>redundancyMeasure</th>\n",
       "      <th>...</th>\n",
       "      <th>lengthyDomain</th>\n",
       "      <th>hyperlinkToAllWordsRatio</th>\n",
       "      <th>isFrontPageNews</th>\n",
       "      <th>alphanumCharCount</th>\n",
       "      <th>linksCount</th>\n",
       "      <th>wordCount</th>\n",
       "      <th>parametrizedLinkRatio</th>\n",
       "      <th>spellingErrorsRatio</th>\n",
       "      <th>label</th>\n",
       "      <th>id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>http://www.polyvore.com/cgi/home?id=1389651</td>\n",
       "      <td>{\"title\":\"Join Polyvore and follow the latest ...</td>\n",
       "      <td>?</td>\n",
       "      <td>?</td>\n",
       "      <td>1.916667</td>\n",
       "      <td>0.047619</td>\n",
       "      <td>0.007937</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.803797</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>34</td>\n",
       "      <td>0</td>\n",
       "      <td>682</td>\n",
       "      <td>126</td>\n",
       "      <td>1</td>\n",
       "      <td>0.531746</td>\n",
       "      <td>0.142857</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3711</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>http://www.youtube.com/watch?v=ippMPPu6gh4</td>\n",
       "      <td>{\"body\":\"Speed Air Man--David Belle david bell...</td>\n",
       "      <td>?</td>\n",
       "      <td>?</td>\n",
       "      <td>1.257576</td>\n",
       "      <td>0.141026</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.142857</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>3008</td>\n",
       "      <td>78</td>\n",
       "      <td>1</td>\n",
       "      <td>0.628205</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>7222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>http://www.musingsofahousewife.com/2011/03/tri...</td>\n",
       "      <td>{\"title\":\"Chicken Gruyere Recipe \",\"body\":\"Chi...</td>\n",
       "      <td>science_technology</td>\n",
       "      <td>0.386685</td>\n",
       "      <td>2.024000</td>\n",
       "      <td>0.630350</td>\n",
       "      <td>0.284047</td>\n",
       "      <td>0.233463</td>\n",
       "      <td>0.202335</td>\n",
       "      <td>0.443409</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>17</td>\n",
       "      <td>0</td>\n",
       "      <td>11008</td>\n",
       "      <td>257</td>\n",
       "      <td>5</td>\n",
       "      <td>0.466926</td>\n",
       "      <td>0.062500</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3964</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>http://www.thelittleteochew.com/2011/07/ikan-b...</td>\n",
       "      <td>{\"title\":\"Ikan Bilis With Tempeh Groundnuts \",...</td>\n",
       "      <td>recreation</td>\n",
       "      <td>0.475039</td>\n",
       "      <td>1.665254</td>\n",
       "      <td>0.419580</td>\n",
       "      <td>0.199301</td>\n",
       "      <td>0.097902</td>\n",
       "      <td>0.066434</td>\n",
       "      <td>0.472649</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>28</td>\n",
       "      <td>0</td>\n",
       "      <td>4585</td>\n",
       "      <td>286</td>\n",
       "      <td>5</td>\n",
       "      <td>0.244755</td>\n",
       "      <td>0.085868</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3697</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>http://recipes.wuzzle.org/index.php/72</td>\n",
       "      <td>{\"url\":\"recipes wuzzle org index php 72\",\"titl...</td>\n",
       "      <td>computer_internet</td>\n",
       "      <td>0.535009</td>\n",
       "      <td>0.181818</td>\n",
       "      <td>0.036364</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.292614</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>1745</td>\n",
       "      <td>55</td>\n",
       "      <td>1</td>\n",
       "      <td>0.072727</td>\n",
       "      <td>0.115044</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4321</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 27 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 url  \\\n",
       "0        http://www.polyvore.com/cgi/home?id=1389651   \n",
       "1         http://www.youtube.com/watch?v=ippMPPu6gh4   \n",
       "2  http://www.musingsofahousewife.com/2011/03/tri...   \n",
       "3  http://www.thelittleteochew.com/2011/07/ikan-b...   \n",
       "4             http://recipes.wuzzle.org/index.php/72   \n",
       "\n",
       "                                  webpageDescription    alchemy_category  \\\n",
       "0  {\"title\":\"Join Polyvore and follow the latest ...                   ?   \n",
       "1  {\"body\":\"Speed Air Man--David Belle david bell...                   ?   \n",
       "2  {\"title\":\"Chicken Gruyere Recipe \",\"body\":\"Chi...  science_technology   \n",
       "3  {\"title\":\"Ikan Bilis With Tempeh Groundnuts \",...          recreation   \n",
       "4  {\"url\":\"recipes wuzzle org index php 72\",\"titl...   computer_internet   \n",
       "\n",
       "  alchemy_category_score  avgLinkWordLength  AvglinkWithOneCommonWord  \\\n",
       "0                      ?           1.916667                  0.047619   \n",
       "1                      ?           1.257576                  0.141026   \n",
       "2               0.386685           2.024000                  0.630350   \n",
       "3               0.475039           1.665254                  0.419580   \n",
       "4               0.535009           0.181818                  0.036364   \n",
       "\n",
       "   AvglinkWithTwoCommonWord  AvglinkWithThreeCommonWord  \\\n",
       "0                  0.007937                    0.000000   \n",
       "1                  0.000000                    0.000000   \n",
       "2                  0.284047                    0.233463   \n",
       "3                  0.199301                    0.097902   \n",
       "4                  0.000000                    0.000000   \n",
       "\n",
       "   AvglinkWithFourCommonWord  redundancyMeasure  ...  lengthyDomain  \\\n",
       "0                   0.000000           0.803797  ...              0   \n",
       "1                   0.000000           1.142857  ...              0   \n",
       "2                   0.202335           0.443409  ...              1   \n",
       "3                   0.066434           0.472649  ...              1   \n",
       "4                   0.000000           0.292614  ...              0   \n",
       "\n",
       "   hyperlinkToAllWordsRatio  isFrontPageNews  alphanumCharCount  linksCount  \\\n",
       "0                        34                0                682         126   \n",
       "1                        12                0               3008          78   \n",
       "2                        17                0              11008         257   \n",
       "3                        28                0               4585         286   \n",
       "4                         3                0               1745          55   \n",
       "\n",
       "   wordCount parametrizedLinkRatio  spellingErrorsRatio  label    id  \n",
       "0          1              0.531746             0.142857    1.0  3711  \n",
       "1          1              0.628205             0.000000    1.0  7222  \n",
       "2          5              0.466926             0.062500    1.0  3964  \n",
       "3          5              0.244755             0.085868    1.0  3697  \n",
       "4          1              0.072727             0.115044    1.0  4321  \n",
       "\n",
       "[5 rows x 27 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6d150565",
   "metadata": {},
   "source": [
    "## 1.1 Checking for NULL and \"?\" values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "4792a3b7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "url                              0\n",
       "webpageDescription               0\n",
       "alchemy_category                 0\n",
       "alchemy_category_score           0\n",
       "avgLinkWordLength                0\n",
       "AvglinkWithOneCommonWord         0\n",
       "AvglinkWithTwoCommonWord         0\n",
       "AvglinkWithThreeCommonWord       0\n",
       "AvglinkWithFourCommonWord        0\n",
       "redundancyMeasure                0\n",
       "embedRatio                       0\n",
       "framebased                       0\n",
       "frameTagRatio                    0\n",
       "domainLink                       0\n",
       "tagRatio                         0\n",
       "imageTagRatio                    0\n",
       "isNews                           0\n",
       "lengthyDomain                    0\n",
       "hyperlinkToAllWordsRatio         0\n",
       "isFrontPageNews                  0\n",
       "alphanumCharCount                0\n",
       "linksCount                       0\n",
       "wordCount                        0\n",
       "parametrizedLinkRatio            0\n",
       "spellingErrorsRatio              0\n",
       "label                         1479\n",
       "id                               0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df.isna().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "102d7d26",
   "metadata": {},
   "source": [
    "No Null Values exist"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "8608e5b1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "url                              0\n",
       "webpageDescription               0\n",
       "alchemy_category              2342\n",
       "alchemy_category_score        2342\n",
       "avgLinkWordLength                0\n",
       "AvglinkWithOneCommonWord         0\n",
       "AvglinkWithTwoCommonWord         0\n",
       "AvglinkWithThreeCommonWord       0\n",
       "AvglinkWithFourCommonWord        0\n",
       "redundancyMeasure                0\n",
       "embedRatio                       0\n",
       "framebased                       0\n",
       "frameTagRatio                    0\n",
       "domainLink                       0\n",
       "tagRatio                         0\n",
       "imageTagRatio                    0\n",
       "isNews                        2843\n",
       "lengthyDomain                    0\n",
       "hyperlinkToAllWordsRatio         0\n",
       "isFrontPageNews               1248\n",
       "alphanumCharCount                0\n",
       "linksCount                       0\n",
       "wordCount                        0\n",
       "parametrizedLinkRatio            0\n",
       "spellingErrorsRatio              0\n",
       "label                            0\n",
       "id                               0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(main_df == '?').sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "fcf1c7cd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2843"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(main_df[\"isNews\"] == '?').sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "84b7d2b9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    4552\n",
       "?    2843\n",
       "Name: isNews, dtype: int64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df[\"isNews\"].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5b5fbd28",
   "metadata": {},
   "source": [
    "Assuming the other 2843 rows are not news. Making the '?' values to 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "b72fcaf6",
   "metadata": {},
   "outputs": [],
   "source": [
    "main_df[\"isNews\"].replace(\"?\", 0, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "9bd1e418",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    4552\n",
       "0    2843\n",
       "Name: isNews, dtype: int64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df[\"isNews\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "442d357d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1248"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(main_df[\"isFrontPageNews\"] == '?').sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "98bd660f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    5853\n",
       "?    1248\n",
       "1     294\n",
       "Name: isFrontPageNews, dtype: int64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df[\"isFrontPageNews\"].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "aa09fed1",
   "metadata": {},
   "source": [
    "Assuming they are not front page news. Changing the \"?\" to 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "f72bfbd5",
   "metadata": {},
   "outputs": [],
   "source": [
    "main_df[\"isFrontPageNews\"].replace(\"?\", 0, inplace = True)\n",
    "main_df[\"isFrontPageNews\"] = pd.to_numeric(main_df[\"isFrontPageNews\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "00de72db",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    7101\n",
       "1     294\n",
       "Name: isFrontPageNews, dtype: int64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df[\"isFrontPageNews\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "83581cfe",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2342"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(main_df[\"alchemy_category_score\"] == '?').sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "647979c3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "?            2342\n",
       "0.85           82\n",
       "0.0784091      40\n",
       "0.849999       12\n",
       "0.535009       10\n",
       "             ... \n",
       "0.181393        1\n",
       "0.640928        1\n",
       "0.638953        1\n",
       "0.881255        1\n",
       "0.848075        1\n",
       "Name: alchemy_category_score, Length: 4806, dtype: int64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df[\"alchemy_category_score\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "bee56e08",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    5053.000000\n",
       "mean        0.603334\n",
       "std         0.212864\n",
       "min         0.070833\n",
       "25%         0.452424\n",
       "50%         0.625616\n",
       "75%         0.780851\n",
       "max         0.999426\n",
       "Name: alchemy_category_score, dtype: float64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df[\"alchemy_category_score\"] = main_df[\"alchemy_category_score\"].replace('?', np.nan)\n",
    "main_df[\"alchemy_category_score\"] = pd.to_numeric(main_df[\"alchemy_category_score\"])\n",
    "main_df[\"alchemy_category_score\"].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "a4d226b3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2342"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df[\"alchemy_category_score\"].isna().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d8ac839b",
   "metadata": {},
   "source": [
    "Mean and Median are almost same. So, replacing nan values with Median"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "74c7f478",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    7395.000000\n",
       "mean        0.610391\n",
       "std         0.176257\n",
       "min         0.070833\n",
       "25%         0.538758\n",
       "50%         0.625616\n",
       "75%         0.708279\n",
       "max         0.999426\n",
       "Name: alchemy_category_score, dtype: float64"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df[\"alchemy_category_score\"].fillna(main_df[\"alchemy_category_score\"].median(), inplace=True)\n",
    "main_df[\"alchemy_category_score\"].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "db785161",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count     7395\n",
       "unique      14\n",
       "top          ?\n",
       "freq      2342\n",
       "Name: alchemy_category, dtype: object"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df[\"alchemy_category\"].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "3da35c1d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "?                     2342\n",
       "recreation            1229\n",
       "arts_entertainment     941\n",
       "business               880\n",
       "health                 506\n",
       "sports                 380\n",
       "culture_politics       343\n",
       "computer_internet      296\n",
       "science_technology     289\n",
       "gaming                  76\n",
       "religion                72\n",
       "law_crime               31\n",
       "unknown                  6\n",
       "weather                  4\n",
       "Name: alchemy_category, dtype: int64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df[\"alchemy_category\"].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c8becfa0",
   "metadata": {},
   "source": [
    "Changing '?' to a new unknown"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "ff32c045",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "main_df[\"alchemy_category\"].replace(\"?\", \"unknown\", inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "f033d4d6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "url                           0\n",
       "webpageDescription            0\n",
       "alchemy_category              0\n",
       "alchemy_category_score        0\n",
       "avgLinkWordLength             0\n",
       "AvglinkWithOneCommonWord      0\n",
       "AvglinkWithTwoCommonWord      0\n",
       "AvglinkWithThreeCommonWord    0\n",
       "AvglinkWithFourCommonWord     0\n",
       "redundancyMeasure             0\n",
       "embedRatio                    0\n",
       "framebased                    0\n",
       "frameTagRatio                 0\n",
       "domainLink                    0\n",
       "tagRatio                      0\n",
       "imageTagRatio                 0\n",
       "isNews                        0\n",
       "lengthyDomain                 0\n",
       "hyperlinkToAllWordsRatio      0\n",
       "isFrontPageNews               0\n",
       "alphanumCharCount             0\n",
       "linksCount                    0\n",
       "wordCount                     0\n",
       "parametrizedLinkRatio         0\n",
       "spellingErrorsRatio           0\n",
       "label                         0\n",
       "id                            0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(main_df == '?').sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c2b8a231",
   "metadata": {},
   "source": [
    "----------------- all '?' and null values are removed ---------------"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e84aee38",
   "metadata": {},
   "source": [
    "## 1.2 Removing useless columns."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "460464c4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2160x2160 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(30,30))\n",
    "sns.heatmap(main_df.corr(), vmin=-1, cmap=\"coolwarm\", annot=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "23a7e264",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    7101\n",
       "1     294\n",
       "Name: isFrontPageNews, dtype: int64"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df[\"isFrontPageNews\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "c54ec933",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    7395\n",
       "Name: framebased, dtype: int64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df[\"framebased\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "13e54e05",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    7238\n",
       "1     157\n",
       "Name: domainLink, dtype: int64"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df[\"domainLink\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "396d4c01",
   "metadata": {},
   "outputs": [],
   "source": [
    "columns_to_be_deleted = [\"framebased\", \"domainLink\"]\n",
    "main_df.drop(axis=\"columns\", labels=columns_to_be_deleted, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "2b49d12e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>url</th>\n",
       "      <th>webpageDescription</th>\n",
       "      <th>alchemy_category</th>\n",
       "      <th>alchemy_category_score</th>\n",
       "      <th>avgLinkWordLength</th>\n",
       "      <th>AvglinkWithOneCommonWord</th>\n",
       "      <th>AvglinkWithTwoCommonWord</th>\n",
       "      <th>AvglinkWithThreeCommonWord</th>\n",
       "      <th>AvglinkWithFourCommonWord</th>\n",
       "      <th>redundancyMeasure</th>\n",
       "      <th>...</th>\n",
       "      <th>lengthyDomain</th>\n",
       "      <th>hyperlinkToAllWordsRatio</th>\n",
       "      <th>isFrontPageNews</th>\n",
       "      <th>alphanumCharCount</th>\n",
       "      <th>linksCount</th>\n",
       "      <th>wordCount</th>\n",
       "      <th>parametrizedLinkRatio</th>\n",
       "      <th>spellingErrorsRatio</th>\n",
       "      <th>label</th>\n",
       "      <th>id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>http://www.polyvore.com/cgi/home?id=1389651</td>\n",
       "      <td>{\"title\":\"Join Polyvore and follow the latest ...</td>\n",
       "      <td>unknown</td>\n",
       "      <td>0.625616</td>\n",
       "      <td>1.916667</td>\n",
       "      <td>0.047619</td>\n",
       "      <td>0.007937</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.803797</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>34</td>\n",
       "      <td>0</td>\n",
       "      <td>682</td>\n",
       "      <td>126</td>\n",
       "      <td>1</td>\n",
       "      <td>0.531746</td>\n",
       "      <td>0.142857</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3711</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>http://www.youtube.com/watch?v=ippMPPu6gh4</td>\n",
       "      <td>{\"body\":\"Speed Air Man--David Belle david bell...</td>\n",
       "      <td>unknown</td>\n",
       "      <td>0.625616</td>\n",
       "      <td>1.257576</td>\n",
       "      <td>0.141026</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.142857</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>3008</td>\n",
       "      <td>78</td>\n",
       "      <td>1</td>\n",
       "      <td>0.628205</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>7222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>http://www.musingsofahousewife.com/2011/03/tri...</td>\n",
       "      <td>{\"title\":\"Chicken Gruyere Recipe \",\"body\":\"Chi...</td>\n",
       "      <td>science_technology</td>\n",
       "      <td>0.386685</td>\n",
       "      <td>2.024000</td>\n",
       "      <td>0.630350</td>\n",
       "      <td>0.284047</td>\n",
       "      <td>0.233463</td>\n",
       "      <td>0.202335</td>\n",
       "      <td>0.443409</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>17</td>\n",
       "      <td>0</td>\n",
       "      <td>11008</td>\n",
       "      <td>257</td>\n",
       "      <td>5</td>\n",
       "      <td>0.466926</td>\n",
       "      <td>0.062500</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3964</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>http://www.thelittleteochew.com/2011/07/ikan-b...</td>\n",
       "      <td>{\"title\":\"Ikan Bilis With Tempeh Groundnuts \",...</td>\n",
       "      <td>recreation</td>\n",
       "      <td>0.475039</td>\n",
       "      <td>1.665254</td>\n",
       "      <td>0.419580</td>\n",
       "      <td>0.199301</td>\n",
       "      <td>0.097902</td>\n",
       "      <td>0.066434</td>\n",
       "      <td>0.472649</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>28</td>\n",
       "      <td>0</td>\n",
       "      <td>4585</td>\n",
       "      <td>286</td>\n",
       "      <td>5</td>\n",
       "      <td>0.244755</td>\n",
       "      <td>0.085868</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3697</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>http://recipes.wuzzle.org/index.php/72</td>\n",
       "      <td>{\"url\":\"recipes wuzzle org index php 72\",\"titl...</td>\n",
       "      <td>computer_internet</td>\n",
       "      <td>0.535009</td>\n",
       "      <td>0.181818</td>\n",
       "      <td>0.036364</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.292614</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>1745</td>\n",
       "      <td>55</td>\n",
       "      <td>1</td>\n",
       "      <td>0.072727</td>\n",
       "      <td>0.115044</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4321</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 25 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 url  \\\n",
       "0        http://www.polyvore.com/cgi/home?id=1389651   \n",
       "1         http://www.youtube.com/watch?v=ippMPPu6gh4   \n",
       "2  http://www.musingsofahousewife.com/2011/03/tri...   \n",
       "3  http://www.thelittleteochew.com/2011/07/ikan-b...   \n",
       "4             http://recipes.wuzzle.org/index.php/72   \n",
       "\n",
       "                                  webpageDescription    alchemy_category  \\\n",
       "0  {\"title\":\"Join Polyvore and follow the latest ...             unknown   \n",
       "1  {\"body\":\"Speed Air Man--David Belle david bell...             unknown   \n",
       "2  {\"title\":\"Chicken Gruyere Recipe \",\"body\":\"Chi...  science_technology   \n",
       "3  {\"title\":\"Ikan Bilis With Tempeh Groundnuts \",...          recreation   \n",
       "4  {\"url\":\"recipes wuzzle org index php 72\",\"titl...   computer_internet   \n",
       "\n",
       "   alchemy_category_score  avgLinkWordLength  AvglinkWithOneCommonWord  \\\n",
       "0                0.625616           1.916667                  0.047619   \n",
       "1                0.625616           1.257576                  0.141026   \n",
       "2                0.386685           2.024000                  0.630350   \n",
       "3                0.475039           1.665254                  0.419580   \n",
       "4                0.535009           0.181818                  0.036364   \n",
       "\n",
       "   AvglinkWithTwoCommonWord  AvglinkWithThreeCommonWord  \\\n",
       "0                  0.007937                    0.000000   \n",
       "1                  0.000000                    0.000000   \n",
       "2                  0.284047                    0.233463   \n",
       "3                  0.199301                    0.097902   \n",
       "4                  0.000000                    0.000000   \n",
       "\n",
       "   AvglinkWithFourCommonWord  redundancyMeasure  ...  lengthyDomain  \\\n",
       "0                   0.000000           0.803797  ...              0   \n",
       "1                   0.000000           1.142857  ...              0   \n",
       "2                   0.202335           0.443409  ...              1   \n",
       "3                   0.066434           0.472649  ...              1   \n",
       "4                   0.000000           0.292614  ...              0   \n",
       "\n",
       "   hyperlinkToAllWordsRatio  isFrontPageNews  alphanumCharCount linksCount  \\\n",
       "0                        34                0                682        126   \n",
       "1                        12                0               3008         78   \n",
       "2                        17                0              11008        257   \n",
       "3                        28                0               4585        286   \n",
       "4                         3                0               1745         55   \n",
       "\n",
       "   wordCount  parametrizedLinkRatio  spellingErrorsRatio  label    id  \n",
       "0          1               0.531746             0.142857    1.0  3711  \n",
       "1          1               0.628205             0.000000    1.0  7222  \n",
       "2          5               0.466926             0.062500    1.0  3964  \n",
       "3          5               0.244755             0.085868    1.0  3697  \n",
       "4          1               0.072727             0.115044    1.0  4321  \n",
       "\n",
       "[5 rows x 25 columns]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "2b8b768d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>url</th>\n",
       "      <th>webpageDescription</th>\n",
       "      <th>alchemy_category</th>\n",
       "      <th>alchemy_category_score</th>\n",
       "      <th>avgLinkWordLength</th>\n",
       "      <th>AvglinkWithOneCommonWord</th>\n",
       "      <th>AvglinkWithTwoCommonWord</th>\n",
       "      <th>AvglinkWithThreeCommonWord</th>\n",
       "      <th>AvglinkWithFourCommonWord</th>\n",
       "      <th>redundancyMeasure</th>\n",
       "      <th>...</th>\n",
       "      <th>lengthyDomain</th>\n",
       "      <th>hyperlinkToAllWordsRatio</th>\n",
       "      <th>isFrontPageNews</th>\n",
       "      <th>alphanumCharCount</th>\n",
       "      <th>linksCount</th>\n",
       "      <th>wordCount</th>\n",
       "      <th>parametrizedLinkRatio</th>\n",
       "      <th>spellingErrorsRatio</th>\n",
       "      <th>label</th>\n",
       "      <th>id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>http://www.polyvore.com/cgi/home?id=1389651</td>\n",
       "      <td>{\"title\":\"Join Polyvore and follow the latest ...</td>\n",
       "      <td>unknown</td>\n",
       "      <td>0.625616</td>\n",
       "      <td>1.916667</td>\n",
       "      <td>0.047619</td>\n",
       "      <td>0.007937</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.803797</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>34</td>\n",
       "      <td>0</td>\n",
       "      <td>682</td>\n",
       "      <td>126</td>\n",
       "      <td>1</td>\n",
       "      <td>0.531746</td>\n",
       "      <td>0.142857</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3711</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>http://www.youtube.com/watch?v=ippMPPu6gh4</td>\n",
       "      <td>{\"body\":\"Speed Air Man--David Belle david bell...</td>\n",
       "      <td>unknown</td>\n",
       "      <td>0.625616</td>\n",
       "      <td>1.257576</td>\n",
       "      <td>0.141026</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.142857</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>3008</td>\n",
       "      <td>78</td>\n",
       "      <td>1</td>\n",
       "      <td>0.628205</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>7222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>http://www.musingsofahousewife.com/2011/03/tri...</td>\n",
       "      <td>{\"title\":\"Chicken Gruyere Recipe \",\"body\":\"Chi...</td>\n",
       "      <td>science_technology</td>\n",
       "      <td>0.386685</td>\n",
       "      <td>2.024000</td>\n",
       "      <td>0.630350</td>\n",
       "      <td>0.284047</td>\n",
       "      <td>0.233463</td>\n",
       "      <td>0.202335</td>\n",
       "      <td>0.443409</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>17</td>\n",
       "      <td>0</td>\n",
       "      <td>11008</td>\n",
       "      <td>257</td>\n",
       "      <td>5</td>\n",
       "      <td>0.466926</td>\n",
       "      <td>0.062500</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3964</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>http://www.thelittleteochew.com/2011/07/ikan-b...</td>\n",
       "      <td>{\"title\":\"Ikan Bilis With Tempeh Groundnuts \",...</td>\n",
       "      <td>recreation</td>\n",
       "      <td>0.475039</td>\n",
       "      <td>1.665254</td>\n",
       "      <td>0.419580</td>\n",
       "      <td>0.199301</td>\n",
       "      <td>0.097902</td>\n",
       "      <td>0.066434</td>\n",
       "      <td>0.472649</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>28</td>\n",
       "      <td>0</td>\n",
       "      <td>4585</td>\n",
       "      <td>286</td>\n",
       "      <td>5</td>\n",
       "      <td>0.244755</td>\n",
       "      <td>0.085868</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3697</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>http://recipes.wuzzle.org/index.php/72</td>\n",
       "      <td>{\"url\":\"recipes wuzzle org index php 72\",\"titl...</td>\n",
       "      <td>computer_internet</td>\n",
       "      <td>0.535009</td>\n",
       "      <td>0.181818</td>\n",
       "      <td>0.036364</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.292614</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>1745</td>\n",
       "      <td>55</td>\n",
       "      <td>1</td>\n",
       "      <td>0.072727</td>\n",
       "      <td>0.115044</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4321</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 25 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 url  \\\n",
       "0        http://www.polyvore.com/cgi/home?id=1389651   \n",
       "1         http://www.youtube.com/watch?v=ippMPPu6gh4   \n",
       "2  http://www.musingsofahousewife.com/2011/03/tri...   \n",
       "3  http://www.thelittleteochew.com/2011/07/ikan-b...   \n",
       "4             http://recipes.wuzzle.org/index.php/72   \n",
       "\n",
       "                                  webpageDescription    alchemy_category  \\\n",
       "0  {\"title\":\"Join Polyvore and follow the latest ...             unknown   \n",
       "1  {\"body\":\"Speed Air Man--David Belle david bell...             unknown   \n",
       "2  {\"title\":\"Chicken Gruyere Recipe \",\"body\":\"Chi...  science_technology   \n",
       "3  {\"title\":\"Ikan Bilis With Tempeh Groundnuts \",...          recreation   \n",
       "4  {\"url\":\"recipes wuzzle org index php 72\",\"titl...   computer_internet   \n",
       "\n",
       "   alchemy_category_score  avgLinkWordLength  AvglinkWithOneCommonWord  \\\n",
       "0                0.625616           1.916667                  0.047619   \n",
       "1                0.625616           1.257576                  0.141026   \n",
       "2                0.386685           2.024000                  0.630350   \n",
       "3                0.475039           1.665254                  0.419580   \n",
       "4                0.535009           0.181818                  0.036364   \n",
       "\n",
       "   AvglinkWithTwoCommonWord  AvglinkWithThreeCommonWord  \\\n",
       "0                  0.007937                    0.000000   \n",
       "1                  0.000000                    0.000000   \n",
       "2                  0.284047                    0.233463   \n",
       "3                  0.199301                    0.097902   \n",
       "4                  0.000000                    0.000000   \n",
       "\n",
       "   AvglinkWithFourCommonWord  redundancyMeasure  ...  lengthyDomain  \\\n",
       "0                   0.000000           0.803797  ...              0   \n",
       "1                   0.000000           1.142857  ...              0   \n",
       "2                   0.202335           0.443409  ...              1   \n",
       "3                   0.066434           0.472649  ...              1   \n",
       "4                   0.000000           0.292614  ...              0   \n",
       "\n",
       "   hyperlinkToAllWordsRatio  isFrontPageNews  alphanumCharCount linksCount  \\\n",
       "0                        34                0                682        126   \n",
       "1                        12                0               3008         78   \n",
       "2                        17                0              11008        257   \n",
       "3                        28                0               4585        286   \n",
       "4                         3                0               1745         55   \n",
       "\n",
       "   wordCount  parametrizedLinkRatio  spellingErrorsRatio  label    id  \n",
       "0          1               0.531746             0.142857    1.0  3711  \n",
       "1          1               0.628205             0.000000    1.0  7222  \n",
       "2          5               0.466926             0.062500    1.0  3964  \n",
       "3          5               0.244755             0.085868    1.0  3697  \n",
       "4          1               0.072727             0.115044    1.0  4321  \n",
       "\n",
       "[5 rows x 25 columns]"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "38000b3e",
   "metadata": {},
   "source": [
    "## 1.3 Skew Checking for Numeric data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "c7860940",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(main_df[\"alchemy_category_score\"], bins = 100)\n",
    "plt.show()\n",
    "\n",
    "main_df[\"alchemy_category_score\"] = np.exp(main_df[\"alchemy_category_score\"])\n",
    "\n",
    "q = main_df[\"alchemy_category_score\"].to_numpy()\n",
    "plt.hist(q, bins = 100)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "590cee1f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(main_df[\"redundancyMeasure\"], bins = 100)\n",
    "plt.show()\n",
    "\n",
    "main_df[\"redundancyMeasure\"] = np.log(main_df[\"redundancyMeasure\"]+1)\n",
    "\n",
    "q = main_df[\"redundancyMeasure\"].to_numpy()\n",
    "plt.hist(q, bins = 100)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "045cf1d8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(main_df[\"linksCount\"], bins = 100)\n",
    "plt.show()\n",
    "\n",
    "main_df[\"linksCount\"] = np.log(main_df[\"linksCount\"]+1)\n",
    "\n",
    "q = main_df[\"linksCount\"].to_numpy()\n",
    "plt.hist(q, bins = 100)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "937c0ef7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(main_df[\"wordCount\"], bins = 100)\n",
    "plt.show()\n",
    "\n",
    "main_df[\"wordCount\"] = np.log(main_df[\"wordCount\"]+1)\n",
    "\n",
    "q = main_df[\"wordCount\"].to_numpy()\n",
    "plt.hist(q, bins = 100)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "c637e8a6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(main_df[\"alphanumCharCount\"], bins = 100)\n",
    "plt.show()\n",
    "\n",
    "main_df[\"alphanumCharCount\"] = np.log(main_df[\"alphanumCharCount\"]+1)\n",
    "\n",
    "q = main_df[\"alphanumCharCount\"].to_numpy()\n",
    "plt.hist(q, bins = 100)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "81a0fc59",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAD4CAYAAAAAczaOAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAAATC0lEQVR4nO3cf6jd933f8eerkqNoSUzs+doISUzKEN1kszjxRdPICNvc1UoyKu8PDwU2i2HQMOqWwsaQVujaPwTuYGUzzAKtySxvWVytbbBo57ZCaygDE/U6sSPLjmYldm1NmnSbLlTZwJ3V9/44Hyen0rn3nitfn3usz/MBX77f8z6fz7nv80H3pe/9nh+pKiRJffix1W5AkjQ5hr4kdcTQl6SOGPqS1BFDX5I6sna1G1jKHXfcUVu2bFntNiTpfeX555//w6qauba+ZOgn+XHgV4ZKHwN+Dniq1bcArwN/r6r+d5tzEHgEuAr8k6r67Va/D3gSWA/8V+ALtcR7Rrds2cLc3NxSbUqShiT5g1H1JS/vVNXZqrq3qu4F7gP+L/BV4ABwsqq2ASfbbZJsB/YAdwO7gCeSrGkPdxjYB2xr26538ZwkScu03Gv69wPfqao/AHYDR1v9KPBgO94NPF1Vb1XVa8A5YEeSDcCtVfVcO7t/amiOJGkClhv6e4CvtOO7quoiQNvf2eobgTeH5pxvtY3t+Nr6dZLsSzKXZG5+fn6ZLUqSFjJ26Cf5APBTwH9ZauiIWi1Sv75YdaSqZqtqdmbmutchJEk3aDln+p8BvlFVl9rtS+2SDW1/udXPA5uH5m0CLrT6phF1SdKELCf0P8+PLu0AHAf2tuO9wDND9T1J1iXZyuAF21PtEtCVJDuTBHh4aI4kaQLGep9+kj8H/G3gHw2VHwOOJXkEeAN4CKCqziQ5BrwMvA3sr6qrbc6j/Ogtm8+2TZI0IZn2r1aenZ0t36cvScuT5Pmqmr227tcwSFJHpv5rGN6NLQd+84fHrz/2uVXsRJKmg2f6ktQRQ1+SOmLoS1JHDH1J6oihL0kdMfQlqSOGviR1xNCXpI4Y+pLUEUNfkjpi6EtSRwx9SeqIoS9JHTH0Jakjhr4kdcTQl6SOGPqS1BFDX5I6YuhLUkcMfUnqyFihn+SjSX41ybeTvJLkryW5PcmJJK+2/W1D4w8mOZfkbJIHhur3JTnd7ns8Sd6LJyVJGm3cM/1/C/xWVf0l4OPAK8AB4GRVbQNOttsk2Q7sAe4GdgFPJFnTHucwsA/Y1rZdK/Q8JEljWDL0k9wKfBr4IkBV/UlVfR/YDRxtw44CD7bj3cDTVfVWVb0GnAN2JNkA3FpVz1VVAU8NzZEkTcA4Z/ofA+aB/5Dkm0l+OcmHgLuq6iJA29/Zxm8E3hyaf77VNrbja+vXSbIvyVySufn5+WU9IUnSwsYJ/bXAJ4HDVfUJ4P/QLuUsYNR1+lqkfn2x6khVzVbV7MzMzBgtSpLGMU7onwfOV9XX2+1fZfCfwKV2yYa2vzw0fvPQ/E3AhVbfNKIuSZqQJUO/qv4X8GaSH2+l+4GXgePA3lbbCzzTjo8De5KsS7KVwQu2p9oloCtJdrZ37Tw8NEeSNAFrxxz3j4EvJ/kA8F3gHzL4D+NYkkeAN4CHAKrqTJJjDP5jeBvYX1VX2+M8CjwJrAeebZskaULGCv2qegGYHXHX/QuMPwQcGlGfA+5ZRn+SpBXkJ3IlqSOGviR1xNCXpI4Y+pLUEUNfkjpi6EtSRwx9SeqIoS9JHTH0Jakjhr4kdcTQl6SOGPqS1BFDX5I6YuhLUkcMfUnqiKEvSR0x9CWpI4a+JHXE0Jekjhj6ktQRQ1+SOjJW6Cd5PcnpJC8kmWu125OcSPJq2982NP5gknNJziZ5YKh+X3ucc0keT5KVf0qSpIUs50z/b1bVvVU1224fAE5W1TbgZLtNku3AHuBuYBfwRJI1bc5hYB+wrW273v1TkCSN691c3tkNHG3HR4EHh+pPV9VbVfUacA7YkWQDcGtVPVdVBTw1NEeSNAHjhn4Bv5Pk+ST7Wu2uqroI0PZ3tvpG4M2huedbbWM7vrZ+nST7kswlmZufnx+zRUnSUtaOOe5TVXUhyZ3AiSTfXmTsqOv0tUj9+mLVEeAIwOzs7MgxkqTlG+tMv6outP1l4KvADuBSu2RD219uw88Dm4embwIutPqmEXVJ0oQsGfpJPpTkI+8cAz8JvAQcB/a2YXuBZ9rxcWBPknVJtjJ4wfZUuwR0JcnO9q6dh4fmSJImYJzLO3cBX23vrlwL/Oeq+q0kvw8cS/II8AbwEEBVnUlyDHgZeBvYX1VX22M9CjwJrAeebZskaUKWDP2q+i7w8RH17wH3LzDnEHBoRH0OuGf5bUqSVoKfyJWkjhj6ktQRQ1+SOmLoS1JHDH1J6oihL0kdMfQlqSOGviR1xNCXpI4Y+pLUEUNfkjpi6EtSRwx9SeqIoS9JHTH0Jakjhr4kdcTQl6SOGPqS1BFDX5I6YuhLUkcMfUnqiKEvSR0ZO/STrEnyzSS/0W7fnuREklfb/rahsQeTnEtyNskDQ/X7kpxu9z2eJCv7dCRJi1nOmf4XgFeGbh8ATlbVNuBku02S7cAe4G5gF/BEkjVtzmFgH7CtbbveVfeSpGUZK/STbAI+B/zyUHk3cLQdHwUeHKo/XVVvVdVrwDlgR5INwK1V9VxVFfDU0BxJ0gSMe6b/b4B/DvzpUO2uqroI0PZ3tvpG4M2hcedbbWM7vrZ+nST7kswlmZufnx+zRUnSUpYM/SR/B7hcVc+P+ZijrtPXIvXri1VHqmq2qmZnZmbG/LGSpKWsHWPMp4CfSvJZ4IPArUn+E3ApyYaqutgu3Vxu488Dm4fmbwIutPqmEXVJ0oQseaZfVQeralNVbWHwAu1/q6q/DxwH9rZhe4Fn2vFxYE+SdUm2MnjB9lS7BHQlyc72rp2Hh+ZIkiZgnDP9hTwGHEvyCPAG8BBAVZ1Jcgx4GXgb2F9VV9ucR4EngfXAs22TJE3IskK/qr4GfK0dfw+4f4Fxh4BDI+pzwD3LbVKStDL8RK4kdcTQl6SOGPqS1BFDX5I6YuhLUkcMfUnqiKEvSR0x9CWpI4a+JHXE0Jekjhj6ktQRQ1+SOmLoS1JHDH1J6oihL0kdMfQlqSOGviR1xNCXpI4Y+pLUEUNfkjpi6EtSR5YM/SQfTHIqyYtJziT5hVa/PcmJJK+2/W1Dcw4mOZfkbJIHhur3JTnd7ns8Sd6bpyVJGmWcM/23gL9VVR8H7gV2JdkJHABOVtU24GS7TZLtwB7gbmAX8ESSNe2xDgP7gG1t27VyT0WStJQlQ78GftBu3tK2AnYDR1v9KPBgO94NPF1Vb1XVa8A5YEeSDcCtVfVcVRXw1NAcSdIEjHVNP8maJC8Al4ETVfV14K6qugjQ9ne24RuBN4emn2+1je342vqon7cvyVySufn5+WU8HUnSYsYK/aq6WlX3ApsYnLXfs8jwUdfpa5H6qJ93pKpmq2p2ZmZmnBYlSWNY1rt3qur7wNcYXIu/1C7Z0PaX27DzwOahaZuAC62+aURdkjQh47x7ZybJR9vxeuAngG8Dx4G9bdhe4Jl2fBzYk2Rdkq0MXrA91S4BXUmys71r5+GhOZKkCVg7xpgNwNH2DpwfA45V1W8keQ44luQR4A3gIYCqOpPkGPAy8Dawv6qutsd6FHgSWA882zZJ0oQsGfpV9S3gEyPq3wPuX2DOIeDQiPocsNjrAZKk95CfyJWkjhj6ktQRQ1+SOmLoS1JHDH1J6oihL0kdMfQlqSOGviR1xNCXpI4Y+pLUEUNfkjpi6EtSRwx9SeqIoS9JHTH0Jakjhr4kdcTQl6SOGPqS1BFDX5I6YuhLUkcMfUnqiKEvSR1ZMvSTbE7yu0leSXImyRda/fYkJ5K82va3Dc05mORckrNJHhiq35fkdLvv8SR5b56WJGmUcc703wb+aVX9ZWAnsD/JduAAcLKqtgEn223afXuAu4FdwBNJ1rTHOgzsA7a1bdcKPhdJ0hKWDP2qulhV32jHV4BXgI3AbuBoG3YUeLAd7waerqq3quo14BywI8kG4Naqeq6qCnhqaI4kaQKWdU0/yRbgE8DXgbuq6iIM/mMA7mzDNgJvDk0732ob2/G19VE/Z1+SuSRz8/Pzy2lRkrSIsUM/yYeBXwN+pqr+eLGhI2q1SP36YtWRqpqtqtmZmZlxW5QkLWGs0E9yC4PA/3JV/XorX2qXbGj7y61+Htg8NH0TcKHVN42oS5ImZJx37wT4IvBKVf3S0F3Hgb3teC/wzFB9T5J1SbYyeMH2VLsEdCXJzvaYDw/NkSRNwNoxxnwK+AfA6SQvtNq/AB4DjiV5BHgDeAigqs4kOQa8zOCdP/ur6mqb9yjwJLAeeLZtkqQJWTL0q+q/M/p6PMD9C8w5BBwaUZ8D7llOg5KkleMnciWpI4a+JHXE0Jekjhj6ktQRQ1+SOmLoS1JHDH1J6oihL0kdMfQlqSOGviR1xNCXpI4Y+pLUEUNfkjpi6EtSRwx9SeqIoS9JHTH0Jakjhr4kdcTQl6SOGPqS1BFDX5I6smToJ/lSkstJXhqq3Z7kRJJX2/62ofsOJjmX5GySB4bq9yU53e57PElW/ulIkhYzzpn+k8Cua2oHgJNVtQ042W6TZDuwB7i7zXkiyZo25zCwD9jWtmsfU5L0Hlsy9Kvq94A/uqa8Gzjajo8CDw7Vn66qt6rqNeAcsCPJBuDWqnquqgp4amiOJGlCbvSa/l1VdRGg7e9s9Y3Am0PjzrfaxnZ8bX2kJPuSzCWZm5+fv8EWJUnXWukXckddp69F6iNV1ZGqmq2q2ZmZmRVrTpJ6d6Ohf6ldsqHtL7f6eWDz0LhNwIVW3zSiLkmaoBsN/ePA3na8F3hmqL4nybokWxm8YHuqXQK6kmRne9fOw0NzJEkTsnapAUm+AvwN4I4k54F/CTwGHEvyCPAG8BBAVZ1Jcgx4GXgb2F9VV9tDPcrgnUDrgWfbJkmaoCVDv6o+v8Bd9y8w/hBwaER9DrhnWd1JklaUn8iVpI4Y+pLUEUNfkjpi6EtSRwx9SeqIoS9JHTH0Jakjhr4kdcTQl6SOGPqS1JElv4bhZrHlwG/+8Pj1xz63ip1I0urxTF+SOmLoS1JHDH1J6oihL0kdMfQlqSOGviR1xNCXpI4Y+pLUEUNfkjpi6EtSRwx9SerIxEM/ya4kZ5OcS3Jg0j9fkno20dBPsgb4d8BngO3A55Nsn2QPktSzSX/L5g7gXFV9FyDJ08Bu4OVJNjH8jZsL8Zs4Jd2MJh36G4E3h26fB/7qtYOS7AP2tZs/SHL2Bn/eHcAf3sjE/OIN/sQbc8N9Tph9riz7XDnvhx5hsn3+hVHFSYd+RtTqukLVEeDIu/5hyVxVzb7bx3mv2efKss+V9X7o8/3QI0xHn5N+Ifc8sHno9ibgwoR7kKRuTTr0fx/YlmRrkg8Ae4DjE+5Bkro10cs7VfV2kp8GfhtYA3ypqs68hz/yXV8imhD7XFn2ubLeD32+H3qEKegzVdddUpck3aT8RK4kdcTQl6SO3JShP81f9ZDk9SSnk7yQZK7Vbk9yIsmrbX/bKvT1pSSXk7w0VFuwryQH2/qeTfLAKvf580n+Z1vTF5J8dgr63Jzkd5O8kuRMki+0+lSt6SJ9TtWaJvlgklNJXmx9/kKrT9t6LtTn9KxnVd1UG4MXiL8DfAz4APAisH21+xrq73Xgjmtq/wo40I4PAL+4Cn19Gvgk8NJSfTH4Co0XgXXA1rbea1axz58H/tmIsavZ5wbgk+34I8D/aP1M1Zou0udUrSmDz/h8uB3fAnwd2DmF67lQn1Oznjfjmf4Pv+qhqv4EeOerHqbZbuBoOz4KPDjpBqrq94A/uqa8UF+7gaer6q2qeg04x2DdV6vPhaxmnxer6hvt+ArwCoNPpE/Vmi7S50JWq8+qqh+0m7e0rZi+9Vyoz4VMvM+bMfRHfdXDYv+IJ62A30nyfPu6CYC7quoiDH4JgTtXrbs/a6G+pnGNfzrJt9rln3f+xJ+KPpNsAT7B4Kxvatf0mj5hytY0yZokLwCXgRNVNZXruUCfMCXreTOG/lhf9bCKPlVVn2TwTaP7k3x6tRu6AdO2xoeBvwjcC1wE/nWrr3qfST4M/BrwM1X1x4sNHVGbWK8j+py6Na2qq1V1L4NP8u9Ics8iw6etz6lZz5sx9Kf6qx6q6kLbXwa+yuBPuUtJNgC0/eXV6/DPWKivqVrjqrrUftH+FPj3/OjP41XtM8ktDIL0y1X16608dWs6qs9pXdPW2/eBrwG7mML1fMdwn9O0njdj6E/tVz0k+VCSj7xzDPwk8BKD/va2YXuBZ1anw+ss1NdxYE+SdUm2AtuAU6vQH/DDX/Z3/F0Gawqr2GeSAF8EXqmqXxq6a6rWdKE+p21Nk8wk+Wg7Xg/8BPBtpm89R/Y5Vev5Xr+avRob8FkG70L4DvCzq93PUF8fY/BK/YvAmXd6A/48cBJ4te1vX4XevsLgz87/x+Ds45HF+gJ+tq3vWeAzq9znfwROA99i8Eu0YQr6/OsM/kz/FvBC2z47bWu6SJ9TtabAXwG+2fp5Cfi5Vp+29Vyoz6lZT7+GQZI6cjNe3pEkLcDQl6SOGPqS1BFDX5I6YuhLUkcMfUnqiKEvSR35/4JdG5XLQgHOAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(main_df[\"avgLinkWordLength\"], bins = 100)\n",
    "plt.show()\n",
    "\n",
    "main_df[\"avgLinkWordLength\"] = np.log(main_df[\"avgLinkWordLength\"]+1)\n",
    "\n",
    "q = main_df[\"avgLinkWordLength\"].to_numpy()\n",
    "plt.hist(q, bins = 100)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "e81c8e99",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(main_df[\"AvglinkWithOneCommonWord\"], bins = 100)\n",
    "plt.show()\n",
    "\n",
    "main_df[\"AvglinkWithOneCommonWord\"] = np.exp(main_df[\"AvglinkWithOneCommonWord\"]+1)\n",
    "\n",
    "q = main_df[\"AvglinkWithOneCommonWord\"].to_numpy()\n",
    "plt.hist(q, bins = 100)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "3fe5c349",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(main_df[\"AvglinkWithTwoCommonWord\"], bins = 100)\n",
    "plt.show()\n",
    "\n",
    "main_df[\"AvglinkWithTwoCommonWord\"] = np.log(main_df[\"AvglinkWithTwoCommonWord\"]+1)\n",
    "\n",
    "q = main_df[\"AvglinkWithTwoCommonWord\"].to_numpy()\n",
    "plt.hist(q, bins = 100)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "72b97fe9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(main_df[\"AvglinkWithThreeCommonWord\"], bins = 100)\n",
    "plt.show()\n",
    "\n",
    "main_df[\"AvglinkWithThreeCommonWord\"] = np.log(main_df[\"AvglinkWithThreeCommonWord\"]+1)\n",
    "\n",
    "q = main_df[\"AvglinkWithThreeCommonWord\"].to_numpy()\n",
    "plt.hist(q, bins = 100)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "9fc69277",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(main_df[\"AvglinkWithFourCommonWord\"], bins = 100)\n",
    "plt.show()\n",
    "\n",
    "main_df[\"AvglinkWithFourCommonWord\"] = np.log(main_df[\"AvglinkWithFourCommonWord\"]+1)\n",
    "\n",
    "q = main_df[\"AvglinkWithFourCommonWord\"].to_numpy()\n",
    "plt.hist(q, bins = 100)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dc4b38a8",
   "metadata": {},
   "source": [
    "## 1.4 Outlier removal"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "72c1ca97",
   "metadata": {},
   "outputs": [],
   "source": [
    "class OutlierRemoval: \n",
    "    def __init__(self, lower_quartile, upper_quartile):\n",
    "        self.lower_whisker = lower_quartile - 1.5*(upper_quartile - lower_quartile)\n",
    "        self.upper_whisker = upper_quartile + 1.5*(upper_quartile - lower_quartile)\n",
    "    def removeOutlier(self, x):\n",
    "        return (x if x <= self.upper_whisker and x >= self.lower_whisker else (self.lower_whisker if x < self.lower_whisker else (self.upper_whisker)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "837c7d8a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['url', 'webpageDescription', 'alchemy_category',\n",
      "       'alchemy_category_score', 'avgLinkWordLength',\n",
      "       'AvglinkWithOneCommonWord', 'AvglinkWithTwoCommonWord',\n",
      "       'AvglinkWithThreeCommonWord', 'AvglinkWithFourCommonWord',\n",
      "       'redundancyMeasure', 'embedRatio', 'frameTagRatio', 'tagRatio',\n",
      "       'imageTagRatio', 'isNews', 'lengthyDomain', 'hyperlinkToAllWordsRatio',\n",
      "       'isFrontPageNews', 'alphanumCharCount', 'linksCount', 'wordCount',\n",
      "       'parametrizedLinkRatio', 'spellingErrorsRatio', 'label', 'id'],\n",
      "      dtype='object')\n"
     ]
    }
   ],
   "source": [
    "print(main_df.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "96321819",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "numeric_col = main_df[\"alchemy_category_score\"]\n",
    "sns.boxplot(x=numeric_col)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "9ede0c2e",
   "metadata": {},
   "outputs": [],
   "source": [
    "outlier_remover = OutlierRemoval(numeric_col.quantile(0.25), numeric_col.quantile(0.75))\n",
    "main_df[\"alchemy_category_score\"] = numeric_col.apply(outlier_remover.removeOutlier)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "ae52d1af",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.boxplot(x=main_df[\"alchemy_category_score\"])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "18ad5184",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "numeric_col = main_df[\"avgLinkWordLength\"]\n",
    "sns.boxplot(x=numeric_col)\n",
    "plt.show()\n",
    "\n",
    "outlier_remover_1 = OutlierRemoval(numeric_col.quantile(0.25), numeric_col.quantile(0.75))\n",
    "main_df[\"avgLinkWordLength\"] = numeric_col.apply(outlier_remover_1.removeOutlier)\n",
    "\n",
    "sns.boxplot(x=main_df[\"avgLinkWordLength\"])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "9a4571c5",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWAAAAEGCAYAAABbzE8LAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAAASlklEQVR4nO3dfZBV9X3H8c+XuwjEBa3gww5Grs5CwMaMEWKtzqTgQwbWyZg40tJpUFLalI7hwTHjtLIt7MyqtWZUoBq1mgpNmgcbYh2y4DNJ68SHXYSqQOCKK4VCQLTo8rBm4dc/zrmXey/7cJe9u9+zu+/XDOM95/zuOb/f+e393N/93XOPFkIQAKDvDfGuAAAMVgQwADghgAHACQEMAE4IYABwUtGdwmPGjAnpdLqXqgIAA1NTU9MHIYSzi9d3K4DT6bQaGxvLVysAGATM7P321jMFAQBOCGAAcEIAA4ATAhgAnBDAAOCEAAYAJwQwADghgAHACQEMAE4IYABwQgADgBMCGACcEMAA4IQABgAnBDAAOCGAAcAJAQwATghgAHBCAAOAk279P+EGihUrViiTyUiSdu/eLUkaO3bsSeWqq6s1f/78Pq0bgMFjUAZwJpPRxre36NhnzlLq8EFJ0t7WwlOROvyhR9UADCKDMoAl6dhnztKRiTUasbVBknRkYk3B9ux6AOgtzAEDgBMCGACcEMAA4IQABgAnBDAAOCGAAcAJAQwATghgAHBCAAOAEwIYAJwQwADghAAGACcEMAA4IYABwAkBDABOCGAAcEIAA4ATAhgAnBDAAOCEAAYAJwQwADghgAHACQEMAE4IYABwQgADgBMCGACcEMAA4IQABgAnBDAAOCGAAcAJAQwATghgAHBCAAOAEwIYAJwQwADghAAGACcEMAA4IYABwAkBDABOCGAAcEIAA4CTfh/AK1as0IoVK7yrISlZdQGQfBXeFeipTCbjXYWcJNUFQPL1+xEwAPRXBDAAOCGAAcAJAQwATghgAHBCAAOAEwIYAJwQwADghAAGACcEMAA4IYABwAkBDABOCGAAcEIAA4ATAhgAnBDAAOCEAAYAJwQwADghgAHACQEMAE4IYABwQgADgBMCGACcEMAA4IQABgAnBDAAOCGAAcAJAQwATghgAHBCAAOAEwIYAJwQwADghAAGACcEMAA4IYABwAkBDABOCGAAcEIAA4CTir44yNSpU3OP582bp1mzZunAgQOqq6vTkiVLNHr06IJlSaqrq9OCBQt07733aufOnfr0008VQpAkVVZWqqWlpeAYmUxG1dXVfdGcDr377rtqaWkpaG9PDBkyRMePH88tZ9udSqV07Nix3PqKigqNGTNGe/fu1ahRo/Txxx9LkqqqqvTRRx/pnHPO0b59+xRC0Lhx43TPPfdIkmpra2Vmmjt3rmpra1VVVaVUKiUzUyqV0u23367ly5ef1EfXXHONHnjgAS1ZskTTpk2TpJP6M18mk9HChQu1bNkyVVdXn1S2sbFRd9xxh+677z5Nnjy5LOeunDpr22C1fPlyrV69WjNnztStt97qXZ1elf96Xr9+fVn33ecj4EceeUSStHLlSr311ltatWrVScvZx/X19dq+fbtaW1tz4SvppPCVpPr6+r5pQCfaq1dP5Idv/v7zw1eS2tratHfvXknKha8k7dmzR0ePHtXOnTt19OhRtba2atu2bblzvGXLFm3evFlLlizRkSNHtGPHDm3fvl3btm3Tli1bVF9f324fPfjgg5Kku+66K3es4v7MV19fr0OHDuX6qLjs0qVLdfz48dybb9J01rbBavXq1ZKkp556yrkm/VuvB3B7o8EnnnhC69atUwhB69atUyaTyS2vXbtWa9euVQhBzc3NJR+nublZmUymfBXvpjvvvNPt2N3V0NCghoaG3HJHbxzNzc3t9lH2zbCtrU0vv/yyDhw4UNCfBw4cyO0jk8nk+rG5uVlNTU0FZV966aXc8VtaWtTU1NRLrT41nbVtsFq+fHnB8kMPPeRUk95XnF/l+nSbZfkjy65MmTIlNDY2dusAHVW4oqJCbW1tqqio0Pnnn69du3apra1NZiZJ6k69soYNG6aJEyd2WS6TyeiTT4MOXTpLI7ZGQXRkYk1BmdM3/lgjT7OSpzU2bdrU7fr2F8V9VLytpqZGDQ0Nuf68/vrrddttt0mS5syZU/BGWllZqaNHj+bKSirYZ2VlpdasWdP7jSrR/fff32HbBqv2XtPl/mieFOVqq5k1hRCmFK/vcgRsZt8ys0Yza9y/f3+3D9yR7Iuura1Nzc3NueX8EVZ3tba2lq1+OKG4j4q3vfDCCwX9+fzzz+e2F3+KaWlpKShbvM9yT+P0VGdtA3qqyy/hQgiPSXpMikbAZTtwL4yA0+m0li1b1mW5hQsXqmnHbzstc3z4KFVfdG5J+5PK/9EkSboaAV977bUFo8Trrrsutz2dTnd7BJwknbUN6CmXy9Bmz56tIUOiQ6dSKdXW1uaWhw4dmnthdldtbW3Z6thdV155pduxu6u757i4j/ItXrxYt9xyS0F/3nzzzbntxX1SV1dXULZ47ryurq7kevWFzto2WN14440FyzNnznSqSf/X6wHc3nzJ3LlzNX36dJmZpk+frurq6tzyjBkzNGPGDJmZ0ul0ycdJp9Oul6HdfffdbsfurpqaGtXUnJjz7mjUmU6n2+2j7KeUiooKTZs2TaNHjy7oz/xLtaqrq3P9mE6nNXny5IKyV199de74lZWVibsMrbO2DVYLFiwoWB7Il6EV51e/vwxt3rx5kqKRxSWXXJIbUeQvZx/X1tZq/PjxGjZsWO5FL7UfGJ6j36xyf3wuHnFm959KpQrWV1RU6LzzzpMkjRo1Kre+qqpKw4cP1wUXXKDhw4dr2LBhmjBhQu4cT5o0SRdffLHq6uo0YsQIXXTRRRo/frwmTJigSZMmqba2tt0+WrRokaRo9JtV3J/5amtrdfrpp+f6qLjs0qVLNWTIkMSNfrM6a9tglR0FM/rtmV6/CqK3LVy4UJJKnqvNPqdpx291ZGJNh1dBjNjaoMndmAM+1boAGPhO+SoIAEDvIIABwAkBDABOCGAAcEIAA4ATAhgAnBDAAOCEAAYAJwQwADghgAHACQEMAE4IYABwQgADgBMCGACcEMAA4IQABgAnBDAAOCGAAcAJAQwATghgAHBCAAOAEwIYAJwQwADghAAGACcEMAA4IYABwAkBDABOCGAAcEIAA4ATAhgAnBDAAOCEAAYAJwQwADghgAHACQEMAE4IYABwQgADgBMCGACcVHhXoKeqq6u9q5CTpLoASL5+H8Dz58/3rkJOkuoCIPmYggAAJwQwADghgAHACQEMAE4IYABwQgADgBMCGACcEMAA4IQABgAnBDAAOCGAAcAJAQwATghgAHBCAAOAEwIYAJwQwADghAAGACcEMAA4IYABwAkBDABOCGAAcEIAA4ATAhgAnBDAAOCEAAYAJwQwADghgAHACQEMAE4IYABwQgADgBMCGACcEMAA4IQABgAnBDAAOCGAAcAJAQwATghgAHBCAAOAEwIYAJwQwADgpMK7Al5Shz/UiK0NSh0+IEkasbXhpO3SuQ41AzBYDMoArq6uzj3evbtNkjR2bHHYnltQDgDKbVAG8Pz5872rAADMAQOAFwIYAJwQwADghAAGACcEMAA4IYABwAkBDABOCGAAcEIAA4ATAhgAnBDAAOCEAAYAJwQwADghgAHACQEMAE4IYABwQgADgBMCGACcEMAA4IQABgAnFkIovbDZfknvn+Kxxkj64BSfmxS0ITkGQjtoQzL0RRvGhRDOLl7ZrQDuCTNrDCFM6ZOD9RLakBwDoR20IRk828AUBAA4IYABwElfBvBjfXis3kIbkmMgtIM2JINbG/psDhgAUIgpCABwQgADgJOyB7CZTTez35hZxsz+pp3tZmbL4+3/bWaXlbsOPVVCG6aa2UEz2xj/+3uPenbGzL5vZvvM7O0OtveHfuiqDf2hHz5rZi+b2RYze8fMFrZTJtF9UWIbEt0XZjbczF43s01xG+raKdP3/RBCKNs/SSlJ70q6SNJpkjZJurioTI2ktZJM0hWSXitnHfqoDVMlrfGuaxft+LKkyyS93cH2RPdDiW3oD/1QJemy+PFISdv64WuilDYkui/ic1sZPx4q6TVJV3j3Q7lHwJdLyoQQdoQQPpX0Y0k3FJW5QdKqEHlV0plmVlXmevREKW1IvBDCryR92EmRpPdDKW1IvBDCnhDChvjxJ5K2SBpbVCzRfVFiGxItPrct8eLQ+F/xFQh93g/lDuCxkv4nb3mXTu6oUsp4KrV+fxh/nFlrZr/fN1Urq6T3Q6n6TT+YWVrSFxWNvvL1m77opA1SwvvCzFJmtlHSPknPhxDc+6GizPuzdtYVv8uUUsZTKfXboOi33S1mViPpaUnje7tiZZb0fihFv+kHM6uU9DNJi0IIHxdvbucpieuLLtqQ+L4IIRyTdKmZnSnp52b2+RBC/vcLfd4P5R4B75L02bzl8yX97ymU8dRl/UIIH2c/zoQQGiQNNbMxfVfFskh6P3Spv/SDmQ1VFFw/DCGsbqdI4vuiqzb0l76QpBDC/0laL2l60aY+74dyB/Abksab2YVmdpqkWZKeKSrzjKSb428cr5B0MISwp8z16Iku22Bm55mZxY8vV3QeD/R5TXsm6f3Qpf7QD3H9npC0JYRwfwfFEt0XpbQh6X1hZmfHI1+Z2QhJ10raWlSsz/uhrFMQIYQ2M/u2pGcVXU3w/RDCO2Y2L97+iKQGRd82ZiQdlvTNctahp0psw02S/trM2iQdkTQrxF+jJoWZ/UjRN9NjzGyXpCWKvnjoF/0gldSGxPeDpKskzZb0Vjz/KEl3SrpA6jd9UUobkt4XVZJWmllK0ZvDT0MIa7yziZ8iA4ATfgkHAE4IYABwQgADgBMCGACcEMAA4IQARreZ2VIz+06Z9jXHzP6pHPsq4VhPmtlhMxuZt26ZmYWk/mgAAxsBjJz4AvSB/jeRUXxzpbit0yTt7utKxNejYpAb6C82dMHM0hbd5/VhRb/n/zszeyO+H2pdXrnFFt0j+QVJn8tbv97MpsSPx5hZc/x4jpmtNrN1ZrbdzP4x7znfNLNtZvZLRRf5Z9d/1cxeM7M3zewFMzs3Xr/UonsDrzezHWa2IO85N8d13WRm/2pmI83svfinszKzUWbWnF2W9CNJfxI/nirpFUltefv7hkX3jd1oZo9mg9LMvmdmjVZ0L1kz+wcz2xzX4bvxuifN7Ka8Mi3xf6dadF/df1P0o4aUmd2Xd77/qrv9h/6t3DfjQf/0OUW/+nla0S+aLld0Y5JnzOzLkg4p+kn2FxX9zWyQ1FTCfi+Nn9Mq6TdmtkJR2NVJmizpoKSXJb0Zl/8vRfdoDWb2F5LukHR7vG2iotHqyHhf35M0QdJiSVeFED4ws7NCCJ+Y2XpJ18ftmSXpZyGE38W/lN0u6QYz+z1JfyrpB5JmSJKZTVIUzlfF5R+W9GeSVklaHEL4MA7kF83sC4ruHfB1SRPjOp9Zwjm5XNLnQwjvmdm3FP3c9UtmNkzSK2b2XAjhvRL2gwGAAIYkvR9CeDUewX1FJwKxUtEdrUZK+nkI4bAkmVnx/T068mII4WD8nM2SxkkaI2l9CGF/vP4nioJUim5+8hOL7sF6mqT8IPpFCKFVUquZ7ZN0rqSrJf17COEDSQohZO8d/Lii8H5a0RvLXxbVa7WiYP4DSfmjzmsUvTG8EYf1CEW3LpSkP44Ds0LRz1ovlrRZ0lFJj5vZLyStKeGcvJ4XsF+R9IW80fIZis43ATxIEMCQohGuFI167wkhPJq/0cwWqePb8rXpxFTW8KJtrXmPj+nE31tH+1oh6f4QwjNmNlXS0i72Ze3tK4TwSjy18keSUkW3HJSim+xvkLQyhHA8DlvF+1sZQvjb/MJmdqGk70j6UgjhIzN7UtLw+L4hlysK7lmSvq3oTSF3Tiza+Wl5uzuUv2tJ80MIz3ZwPjDAMQeMfM9K+nOL7vsqMxtrZudI+pWkr5vZiPgKgq/mPadZ0ahRiqYvuvKapKlmNjqel52Zt+0MnfhC7JYS9vWiopHp6Li+Z+VtW6Vovvdfip8UQtipaOri4Xb2d1PcZpnZWWY2TtIoRcF5MJ6Xzk5ZVEo6I7794iJFUy5S4Tm5QfENhNrxrKIb2GTnqyeY2ekltBsDBCNg5IQQnovnQX8djwpbJH0jhLAhnirYKOl9Sf+Z97TvSvqpmc2W9FIJx9hjZksl/VrSHkUj0ewVAUslPWVmuyW9KunCLvb1jpndJemXZnZM0dTJnHjzDyXVKwrh9p77aDvrNptZraTnLLpC4neSbo2nZ96U9I6kHYq+uJOiqZn/MLPhikazt8Xr/zle/7qiUM8f9eZ7XFJa0oZ4pLxf0tc6azMGFu6GhgEpnle9IYQw27suQEcYAWPAia+2mKHo3q5AYjECBgAnfAkHAE4IYABwQgADgBMCGACcEMAA4OT/AdEUwbZK+djZAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "numeric_col = main_df[\"redundancyMeasure\"]\n",
    "sns.boxplot(x=numeric_col)\n",
    "plt.show()\n",
    "\n",
    "outlier_remover_2 = OutlierRemoval(numeric_col.quantile(0.25), numeric_col.quantile(0.75))\n",
    "main_df[\"redundancyMeasure\"] = numeric_col.apply(outlier_remover_2.removeOutlier)\n",
    "\n",
    "sns.boxplot(x=main_df[\"redundancyMeasure\"])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "d066f788",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "numeric_col = main_df[\"embedRatio\"]\n",
    "sns.boxplot(x=numeric_col)\n",
    "plt.show()\n",
    "\n",
    "outlier_remover_3 = OutlierRemoval(numeric_col.quantile(0.25), numeric_col.quantile(0.75))\n",
    "main_df[\"embedRatio\"] = numeric_col.apply(outlier_remover_3.removeOutlier)\n",
    "\n",
    "sns.boxplot(x=main_df[\"embedRatio\"])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "55a0a700",
   "metadata": {},
   "outputs": [],
   "source": [
    "main_df.drop(axis=\"columns\", labels=\"embedRatio\", inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "5f8b3a57",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "numeric_col = main_df[\"frameTagRatio\"]\n",
    "sns.boxplot(x=numeric_col)\n",
    "plt.show()\n",
    "\n",
    "outlier_remover_4 = OutlierRemoval(numeric_col.quantile(0.25), numeric_col.quantile(0.75))\n",
    "main_df[\"frameTagRatio\"] = numeric_col.apply(outlier_remover_4.removeOutlier)\n",
    "\n",
    "sns.boxplot(x=main_df[\"frameTagRatio\"])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "04ecf615",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "numeric_col = main_df[\"tagRatio\"]\n",
    "sns.boxplot(x=numeric_col)\n",
    "plt.show()\n",
    "\n",
    "outlier_remover_5 = OutlierRemoval(numeric_col.quantile(0.25), numeric_col.quantile(0.75))\n",
    "main_df[\"tagRatio\"] = numeric_col.apply(outlier_remover_5.removeOutlier)\n",
    "\n",
    "sns.boxplot(x=main_df[\"tagRatio\"])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "cc377924",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "numeric_col = main_df[\"imageTagRatio\"]\n",
    "sns.boxplot(x=numeric_col)\n",
    "plt.show()\n",
    "\n",
    "outlier_remover_6 = OutlierRemoval(numeric_col.quantile(0.25), numeric_col.quantile(0.75))\n",
    "main_df[\"imageTagRatio\"] = numeric_col.apply(outlier_remover_6.removeOutlier)\n",
    "\n",
    "sns.boxplot(x=main_df[\"imageTagRatio\"])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "108bdcc6",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "numeric_col = main_df[\"lengthyDomain\"]\n",
    "sns.boxplot(x=numeric_col)\n",
    "plt.show()\n",
    "\n",
    "outlier_remover_7 = OutlierRemoval(numeric_col.quantile(0.25), numeric_col.quantile(0.75))\n",
    "main_df[\"lengthyDomain\"] = numeric_col.apply(outlier_remover_7.removeOutlier)\n",
    "\n",
    "sns.boxplot(x=main_df[\"lengthyDomain\"])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "ab27ed13",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "numeric_col = main_df[\"hyperlinkToAllWordsRatio\"]\n",
    "sns.boxplot(x=numeric_col)\n",
    "plt.show()\n",
    "\n",
    "outlier_remover_8 = OutlierRemoval(numeric_col.quantile(0.25), numeric_col.quantile(0.75))\n",
    "main_df[\"hyperlinkToAllWordsRatio\"] = numeric_col.apply(outlier_remover_8.removeOutlier)\n",
    "\n",
    "sns.boxplot(x=main_df[\"hyperlinkToAllWordsRatio\"])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "091e9254",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWAAAAEGCAYAAABbzE8LAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAAANEklEQVR4nO3dfZCddXnG8esiK2NC1QpBancKKW4EClTEFKV2IlSnLU6Vvuioox2wVf9ouwnaTsfp9GWYju3UOu3gapkGbIOtMlMstdVaBielxcqg3UiApAnjgi+wYgiNvJTgS8jdP57fTrZrlj0bztkre/b7mXH25Ow557mf3eXrs8/Z8zuuKgEAlt5x6QEAYKUiwAAQQoABIIQAA0AIAQaAkJHF3Hjt2rW1bt26AY0CAMNp+/btD1fVyXOvX1SA161bp8nJyf5NBQArgO2vHel6TkEAQAgBBoAQAgwAIQQYAEIIMACEEGAACCHAABBCgAEghAADQAgBBoAQAgwAIQQYAEIIMACEEGAACCHAABBCgAEghAADQAgBBoAQAgwAIYt6TzisbBMTE5qamkqP0ZPp6WlJ0ujoaHSOsbExjY+PR2fAsYsAo2dTU1PasXO3nlpzYnqUBa068Kgk6Zvfyf2IrzqwP7ZtLA8EGIvy1JoT9eSZr02PsaDVez4jSdFZZ2YA5sM5YAAIIcAAEEKAASCEAANACAEGgBACDAAhBBgAQggwAIQQYAAIIcAAEEKAASCEAANACAEGgBACDAAhBBgAQggwAIQQYAAIIcAAEEKAASCEAANACAEGgBACDAAhBBgAQggwAIQQYAAIIcAAEEKAASCEAANACAEGgBACDAAhBBgAQggwAIQQYAAIIcAAEEKAASCEAANACAEGgBACDAAhBBgAQggwAIQQYAAIWZIAT0xMaGJiYik2BQB9Nch+jQzkUeeYmppais0AQN8Nsl+cggCAEAIMACEEGABCCDAAhBBgAAghwAAQQoABIIQAA0AIAQaAEAIMACEEGABCCDAAhBBgAAghwAAQQoABIIQAA0AIAQaAEAIMACEEGABCCDAAhBBgAAghwAAQQoABIIQAA0AIAQaAEAIMACEEGABCCDAAhBBgAAghwAAQQoABIIQAA0AIAQaAEAIMACEEGABCCDAAhBBgAAghwAAQQoABIIQAA0AIAQaAkJGl2Mj09LSefPJJbd68eSk2hwGZmprScd+t9BjLxnHffkxTU4/zc7/MTU1NafXq1QN57AWPgG2/y/ak7cl9+/YNZAgAWIkWPAKuqi2StkjShg0bjurwZ3R0VJJ01VVXHc3dcYzYvHmztt+3Nz3GsnHo2c/V2Omn8HO/zA3yNxjOAQNACAEGgBACDAAhBBgAQggwAIQQYAAIIcAAEEKAASCEAANACAEGgBACDAAhBBgAQggwAIQQYAAIIcAAEEKAASCEAANACAEGgBACDAAhBBgAQggwAIQQYAAIIcAAEEKAASCEAANACAEGgBACDAAhBBgAQggwAIQQYAAIIcAAEEKAASCEAANACAEGgBACDAAhBBgAQggwAIQQYAAIIcAAEEKAASCEAANACAEGgJCRpdjI2NjYUmwGAPpukP1akgCPj48vxWYAoO8G2S9OQQBACAEGgBACDAAhBBgAQggwAIQQYAAIIcAAEEKAASCEAANACAEGgBACDAAhBBgAQggwAIQQYAAIIcAAEEKAASCEAANACAEGgBACDAAhBBgAQggwAIQQYAAIIcAAEEKAASCEAANACAEGgBACDAAhBBgAQggwAIQQYAAIIcAAEEKAASCEAANACAEGgBACDAAhBBgAQggwAIQQYAAIIcAAEEKAASBkJD0AlpdVB/Zr9Z7PpMdY0KoD/yNJ0VlXHdgv6ZTY9nHsI8Do2djYWHqEnk1PH5QkjY4mA3jKsvqaYekRYPRsfHw8PQIwVDgHDAAhBBgAQggwAIQQYAAIIcAAEEKAASCEAANACAEGgBACDAAhBBgAQggwAIQQYAAIIcAAEEKAASCEAANACAEGgBACDAAhBBgAQggwAIQQYAAIcVX1fmP7cUn3DG6cqLWSHk4PMQDDul/S8O7bsO6XNLz7ttB+nVZVJ8+9crHvinxPVW1Y5H2WBduTw7hvw7pf0vDu27DulzS8+3a0+8UpCAAIIcAAELLYAG8ZyBTHhmHdt2HdL2l4921Y90sa3n07qv1a1JNwAID+4RQEAIQQYAAI6TnAtr9q+27bO2xPDnKopWT7B21/wvYe27ttX5ieqR9sn9G+VzP/e8z2Fem5+sH2u23vsr3T9vW2n52eqV9sb277tWs5f79s/7Xth2zvnHXdibY/a/vL7ePzkzMerXn27Y3te3bIds9/jrbYI+CLq+q8Ifs7vqsk3VRVZ0p6iaTd4Xn6oqruad+r8yS9TNIBSf+YneqZsz0qaZOkDVV1jqRVkt6cnao/bJ8j6Z2SLlD3s/jzttdnpzpqWyX93Jzr3itpW1Wtl7St/Xs52qrv37edkn5J0q2LeaAVfQrC9nMlbZT0EUmqqu9W1SPRoQbj1ZLuraqvpQfpkxFJq22PSFoj6RvhefrlLEm3V9WBqjoo6T8k/WJ4pqNSVbdK2j/n6kslXdcuXyfpF5Zypn450r5V1e6qWvSrhBcT4JJ0s+3ttt+12A0do06XtE/S39i+w/a1tk9IDzUAb5Z0fXqIfqiqaUkfkPR1SQ9KerSqbs5O1Tc7JW20fZLtNZJeK+lHwjP10ylV9aAktY8vCM8Tt5gAv7Kqzpd0iaTfsL1xQDMtpRFJ50u6uqpeKukJLd9fi47I9vGSXi/phvQs/dDOG14q6Ucl/bCkE2y/LTtVf1TVbkl/Kumzkm6SdKekg9GhMFA9B7iqvtE+PqTuXOIFgxpqCT0g6YGq+kL79yfUBXmYXCLpS1W1Nz1In7xG0leqal9VfU/SjZJ+MjxT31TVR6rq/KraqO7X3C+nZ+qjvbZfKEnt40PheeJ6CrDtE2w/Z+aypJ9R9+vSslZV35R0v+0z2lWvlvTfwZEG4S0aktMPzdclvcL2GttW9z0biidOJcn2C9rHU9U9qTNM37t/lnRZu3yZpH8KznJM6OmVcLZP1+Fn0Eckfbyq3jfIwZaK7fMkXSvpeEn3SXp7VX0rOlSftPOI90s6vaoeTc/TL7avlPQmdb+e3yHpHVX1nexU/WH7c5JOkvQ9Se+pqm3hkY6K7eslXaRumca9kv5Q0icl/b2kU9X9H+kbq2ruE3XHvHn2bb+kCUknS3pE0o6q+tkFH4uXIgNAxor+MzQASCLAABBCgAEghAADQAgBBoAQAozv01a+W/tMb7NUbP+A7b+yfW9bkepW2y+3vW72ilXP4PEvsT3ZVsvbY/sD/Zh7zjauaH82iBWEAGMYXKvu7zDXV9XZki5X9zeaz4jtkbZC2Yckva2qzpJ0jrq/F++3K9QtLIQVhACvcLY/2RZY2jV3kaV2BLnH9nW272rrJs+OxLjtL7V1os9s97nA9m1tcaPbZl5laPty2zfavqmtB/v+Wdv531mX32B7a7u81fbVtm+xfZ/tV7W1WHfPus2LJL1c0u9V1SFJqqr7qupf2kOusn1N27+bba9u93un7f+yfaftf5jZr7bNP7d9i7p1GX5H0vuqak977INV9ZfttqfZ3ta+Ntvaq9dmHuMNc/fP9kW2/92H15/+mDub1K1rcUvbLlYIAoxfraqXSdogaZPtk+Z8/gxJW6rqxyU9JunXZ33u4bZA09WSfrtdt0fSxra40R9I+uNZtz9P3SvYzpX0Jtu9rPT1fEk/Lendkj4l6S8knS3p3PYqxrPVveroqXnuv17Sh9uR8SOSfrldf2NV/URVzawB/Wuz7vNiSa+pqt9Sd8S7fZ7H/pCkj7avzcckfbCH/XmpuqPdH1O3Gt8rq+qD6pbUvLiqLu7hMTAkCDA22b5T0u3qlj6cuwD4/VX1+Xb57yT91KzP3dg+bpe0rl1+nqQb2rnXmVjO2FZVj1bVt9WtuXFaD/N9qrqXa94taW9V3d2OdHfN2ubT+UpV7TjCnOfY/pztuyW9dc6cNzxN0Ge7UNLH2+W/1f//2szni1X1QNuHHeptHzCkCPAKZvsidauLXdiOBO+QNPftfea+Vn32v2fWX3hK3RohkvRHkm5p71bxujmPN3u9htn3mf2Yc7c/c59Dc+5/qN1/l6SX2J7vZ3m+bW6V9JtVda6kK+ds94lZl3epe0eRXszsx0G1/7bagkHH9zAPViACvLI9T9K3qupAO4f7iiPc5lQffp+8t0j6zx4ec7pdvrzHOfbaPqtFdFHvAFFV90qalHRli51sr7d96QJ3fY6kB20/S90R8Hz+TNLv2n5xe+zjbL+nfe42HX47pLfq8Nfmqzoc7UslPauHXXm8zYQVhACvbDdJGrF9l7oj19uPcJvdki5rtzlR3fnep/N+SX9i+/Pq3q+tF++V9GlJ/6buXS4W6x2SfkjSVDulcI0Wfpui35f0BXWLn++Z70ZVdZe6c7bX296tbhnWF7ZPb5L09va1+RVJm9v110h6le0vqnuC8AktbIukf+VJuJWF1dAwL9vrJH26nU4A0GccAQNACEfAABDCETAAhBBgAAghwAAQQoABIIQAA0DI/wFbl97kNtpZDgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "numeric_col = main_df[\"alphanumCharCount\"]\n",
    "sns.boxplot(x=numeric_col)\n",
    "plt.show()\n",
    "\n",
    "outlier_remover_9 = OutlierRemoval(numeric_col.quantile(0.25), numeric_col.quantile(0.75))\n",
    "main_df[\"alphanumCharCount\"] = numeric_col.apply(outlier_remover_9.removeOutlier)\n",
    "\n",
    "sns.boxplot(x=main_df[\"alphanumCharCount\"])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "a9bc3918",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWAAAAEGCAYAAABbzE8LAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAAAQ50lEQVR4nO3df5DU9X3H8ddb7odwiRM5wSHguNKNglPUkBtta+PQKsiZkKQznUmcaYW0TEamPYjSqJ25qRqv7dApTpE/tJa0QJMxbdXMlIYfYqo1mUmTHETBeNSueolHiBCgKkqOO3j3j/1+193l9u72fvBe7p6PGcfbve9+v+87lief+97e98zdBQA49y6IHgAAJisCDABBCDAABCHAABCEAANAkLpqNr7kkks8k8mM0ygAMDHt2bPnl+4+o/z+qgKcyWTU2dk5dlMBwCRgZj8d6H5OQQBAEAIMAEEIMAAEIcAAEIQAA0AQAgwAQQgwAAQhwAAQhAADQBACDABBCDAABCHAABCEAANAEAIMAEEIMAAEIcAAEIQAA0AQAgwAQQgwAASp6nfCAVE2btyoXC43Zvs7ePCgJGn27Nmj3lc2m1VbW9uo94PJhwDjvJDL5fTiy106PW36mOxvyvtvS5J+0Tu6vwJT3j82FuNgkiLAOG+cnjZdJ+fdNib7mnpguySNen/pfoCR4BwwAAQhwAAQhAADQBACDABBCDAABCHAABCEAANAEAIMAEEIMAAEIcAAEIQAA0AQAgwAQQgwAAQhwAAQhAADQBACDABBCDAABCHAABCEAANAEAIMAEEIMAAEIcAAEIQAA0AQAgwAQQgwAAQhwAAQhAADQBACDABBCDAABCHAABCEAANAEAIMAEEIMAAEIcAAEIQAA0AQAgwAQQgwAAQhwAAQhAADQBACDABBCDAABCHAE8DGjRu1cePG6DEwQfH8Gj910QNg9HK5XPQImMB4fo0fVsAAEIQAA0AQAgwAQQgwAAQhwAAQhAADQBACDABBCDAABCHAABCEAANAEAIMAEEIMAAEIcAAEIQAA0AQAgwAQQgwAAQhwAAQhAADQBACDABBCDAABCHAABCEAANAEAIMAEEIMAAEIcAAEIQAA0AQAgwAQQgwAAQhwAAQhAADQBACDABBCDAABCHAABCEAANAEAIMAEEIMAAEIcAAEIQAA0CQunNxkJUrVyqXy2nevHl67LHHqn784sWL1dfXp4aGBj3zzDOjPtbq1au1b98+LVy4UA8//HDJ+1asWKHu7m5ls1lt2rSpcP+6deu0Y8cOLVu2TGvXrh1wP0uWLNGpU6fU2NioXbt2SZJWrVqlrq4uLViwQD09PTp+/Liam5v11FNP6fbbb9ehQ4c0Z84cSVJPT48ymYx6e3t16NChwrFnzpyphoYG9fT0yMzk7iUzX3vttcP8TAIjc+zYMS1atEgzZszQkSNHJEnTp0/XsWPHVF9fryuuuEL33HOP1q9fLzPTQw89pObm5pJ95HI5tbW1adasWWpsbFRHR4eam5uVy+W0Zs0abdiwQdlsdtA5jh49qvb29orHqPSYBx98UPfff/+wti9XzXzVOicr4FwuJ0k6cODAiB7f19cnSTp16tSYHGvfvn2SpL179571vu7u7pL9pHbs2CFJ2rZtW8X9pPP19vYWtunq6pIk7d+/X8ePH5eUf0JIKkS2p6dHPT09heMXx1eSDh8+XHh/eXyBc+HNN9+UpEJ8pXyUpfzfz1dffVUdHR3q6urSK6+8oq1bt561j46ODp08eVKvv/66urq6Ctt0dHTovffeU0dHx5BzbNmyZdBjVHrM/v37h739QHMPd75qjXuAV65cWXL7zjvvrOrxixcvLrm9ZMmSUR1r9erVJbfvvvvuwtsrVqwYcH/r1q0ruX/9+vVn7WfRokUlt2+99VatWrWq4qzl24/GSy+9NGb7AsqloR1KuniR8guWdKEh5Rc0xe9Pt+ns7Czc393dfdbCp9jRo0cLC6GBjlHpMTt37pS7a+fOnUNuX6547qHmG4lxPwVRPnC1q+B09ZsabBU8nGOlq9ZU8Sq4/AmS7q/4D10qXQVX0tvbW1j9ngvpl0kTVS6X0wWnam/1f8Gv3lEu9+6E/tynq99q9PX1aevWrbrrrrskacDVY19fnx544IGS+zo6OrR58+YB97llyxb19/dXPEalx5w5c0aSdPr06SG3L1c+92DzjcSQK2Az+5KZdZpZZ/GXHwBQibtr9+7dhdvli5t0mxMnTpTcN9B2qWeffbbkFFz5MSo9Jo12f3//kNuXK59nsPlGYsgVsLs/LulxSWppaam9JQgkSdlsVhs2bIgeY9ysWbNGe15/K3qMs5y58CJl5146oT/3IzldZmYlpw8zmcxZ8TIzNTU1lUQ4k8lU3Octt9yibdu2FSJcfoxKj9m+fbv6+/tVV1c35PblyucebL6RGPdzwOXfNZw3b15Vj6+vry+53dDQMKpjXXPNNSW3Fy5cWHi7/JOb7q+1tbXk/mXLlp21n3KNjY2aP3/+oNsA54PLLrus6sfU19frjjvuKNxub28fcJvyUxADbZdavny56uo+WDOWH6PSYy64IJ+5KVOmDLl9ufJ5BptvJMY9wMUv5ZJU9cvQyr9kGOxlaMM51iOPPFJyu/hlaOXndtL93XvvvSX3r1279qz9PP/88yW3d+3apUcffbTirOXbjwYvQ8N4mj59+rC2K17AtLa2lrzkK5vNnrXAaW1tVUtLS+H+TCYz6Mu8mpubSxZD5ceo9JilS5fKzLR06dKqX4ZWPPdQ843EOXkZWjp0tavfVLoKHmz1W82x0tVr8eo3lX6yyz/R6R/8smXLKu4nna+xsbGwTboKXrBggS6++GJJKjwJZs2aJUmaM2dO4bXAmUymcH9q5syZhfebWcWPCxgv6Sp4xowZhfvSMNfX1+vKK69Ue3u75s+fr6uvvnrAlWZ7e7umTp2quXPnav78+YVt2tvb1dTUNKzV5fLlywc9RqXHLFiwoOrVb/Hcw52vWlbN60pbWlq8s7NzzIfA6KTfgZ/I5yHTc8An5902JvubemC7JI16f1MPbNcnJvg54Mnw/BpvZrbH3VvK7+dHkQEgCAEGgCAEGACCEGAACEKAASAIAQaAIAQYAIIQYAAIQoABIAgBBoAgBBgAghBgAAhCgAEgCAEGgCAEGACCEGAACEKAASAIAQaAIAQYAIIQYAAIQoABIAgBBoAgBBgAghBgAAhCgAEgCAEGgCAEGACCEGAACEKAASAIAQaAIAQYAIIQYAAIQoABIAgBBoAgBBgAghBgAAhCgAEgCAEGgCB10QNg9LLZbPQImMB4fo0fAjwBtLW1RY+ACYzn1/jhFAQABCHAABCEAANAEAIMAEEIMAAEIcAAEIQAA0AQAgwAQQgwAAQhwAAQhAADQBACDABBCDAABCHAABCEAANAEAIMAEEIMAAEIcAAEIQAA0AQAgwAQQgwAAQhwAAQhAADQBACDABBCDAABCHAABCEAANAEAIMAEEIMAAEIcAAEIQAA0AQAgwAQQgwAAQhwAAQhAADQBACDABBCDAABCHAABCEAANAEAIMAEHqogcAhmvK+8c09cD2MdrXUUka9f6mvH9M0qVjMBEmIwKM80I2mx3T/R082C9Jmj17tPG8dMxnw+RBgHFeaGtrix4BGHOcAwaAIAQYAIIQYAAIQoABIAgBBoAgBBgAghBgAAhCgAEgCAEGgCAEGACCEGAACEKAASAIAQaAIAQYAIIQYAAIQoABIAgBBoAgBBgAghBgAAhCgAEgiLn78Dc2OyLpp+M3TolLJP3yHB2rWsw2MrU6W63OJTHbSNXabJe7+4zyO6sK8LlkZp3u3hI9x0CYbWRqdbZanUtitpGq5dmKcQoCAIIQYAAIUssBfjx6gEEw28jU6my1OpfEbCNVy7MV1Ow5YACY6Gp5BQwAExoBBoAgNRdgM/tHMztsZi9Hz1LMzC4zs+fMrMvMfmJma6JnSpnZhWb2QzN7KZntweiZypnZFDP7sZn9R/Qsxcys28z2m9mLZtYZPU8xM/uImT1pZgeS591vRs8kSWZ2VfL5Sv97x8y+HD1XyszuSv4evGxmT5jZhdEzVVJz54DN7CZJJyRtdfdfj54nZWazJM1y971m9mFJeyR9zt1fCR5NZmaSmtz9hJnVS/qepDXu/t/BoxWY2d2SWiRd5O6fjp4nZWbdklrcvZZetC9JMrMtkr7r7pvMrEHSNHf/v+CxSpjZFEkHJd3g7ufqh7QGm2e28s//q939pJn9q6Tt7r45drKB1dwK2N1fkHQseo5y7n7I3fcmb78rqUvS7Nip8jzvRHKzPvmvZv5lNbM5kj4laVP0LOcLM7tI0k2SviZJ7n6q1uKbuFnSa7UQ3yJ1kqaaWZ2kaZJ+HjxPRTUX4POBmWUkfVzSD4JHKUi+xH9R0mFJu929ZmaT9HeS7pF0JniOgbikZ8xsj5l9KXqYInMlHZH0T8mpm01m1hQ91AC+IOmJ6CFS7n5Q0t9K+pmkQ5LedvdnYqeqjABXycw+JOkpSV9293ei50m5+2l3v07SHEnXm1lNnL4xs09LOuzue6JnqeBGd18oqVXSnySnwGpBnaSFkh51949Lek/SfbEjlUpOi3xG0r9Fz5Iys4slfVbSFZI+KqnJzP4gdqrKCHAVkvOrT0n6hrs/HT3PQJIvU5+XtDR2koIbJX0mOdf6TUm/a2Zfjx3pA+7+8+T/hyV9S9L1sRMV9EjqKfpK5knlg1xLWiXtdfe3ogcpcoukN9z9iLv3SXpa0m8Fz1QRAR6m5BtdX5PU5e4PR89TzMxmmNlHkrenKv8kPBA6VMLd/9zd57h7RvkvV//T3WtiRWJmTck3VJV8eb9EUk28+sbdfyHpTTO7KrnrZknh3/Atc7tq6PRD4meSfsPMpiV/Z29W/vs1NanmAmxmT0j6vqSrzKzHzP44eqbEjZL+UPkVXPrym9uih0rMkvScme2T9CPlzwHX1Mu9atSlkr5nZi9J+qGkb7v7zuCZirVJ+kby53qdpL+KHecDZjZN0mLlV5g1I/mK4UlJeyXtV75xNftjyTX3MjQAmCxqbgUMAJMFAQaAIAQYAIIQYAAIQoABIAgBxrgzsxPJ/z9qZk8Od/th7vtDZvb3ZvZacgWsF8zshtHMO8AxrquhlxxiAqmLHgCTR/JTZ78/xrvdJOkNSR9z9zNmNlfS/DE+xnXKX8lt+xjvF5McK2CcM2aWSa/zbGYrzOxpM9tpZv9rZn8zwPaXmNn3zexTZjYrWd2+mFzn9ZNm9muSbpDU7u5nJMndX3f3byePvzvZ9uX0erXFMyS3/8zMHkjeft7M1iXXVn41OUaDpK9K+nxy7M+P72cJkwkrYES6TvmryvVK+h8z2+jub0qSmV0q6d+Vj+tuM1sraZe7/2VyDdppkn5H0ovufrp8x2b2CUlfVD7QJukHZvZfko4PMVOdu1+fnHK4391vMbO/UP6awX86Fh80kGIFjEjfcfe33f1Xyl/n4PLk/npJ35F0j7vvTu77kaQvJqvVBck1mQfz25K+5e7vJddKflrSJ4cxU/qjtXskZYb9kQAjQIARqbfo7dP64CuyfuUDeGv6zuRC/Tcp/9sX/tnM7pD0E0nXmtlAz2OrcMx+lT7vy39dTTpT8TzAuCDAqEUu6Y8kzTOz+yTJzC5X/rrC/6D8VekWuvtrkjolPZhc+Upm9jEz+6ykFyR9LrkqVpOk35P0XUlvSZppZs1m1ihpOL8e6V1JHx7bDxHgX3jUKHc/bWZfkLTNzN5R/oLkXzGzPuV/Z+AdyaYrJa2XlDOz9yUdlfSV5Hf3bVb+KmeStMndfyxJZvZV5X+byRsa3mU7n5N0X/IbR/7a3f9lLD5GgKuhAUAQTkEAQBACDABBCDAABCHAABCEAANAEAIMAEEIMAAE+X+VkuMBl9PylwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "numeric_col = main_df[\"linksCount\"]\n",
    "sns.boxplot(x=numeric_col)\n",
    "plt.show()\n",
    "\n",
    "outlier_remover_10 = OutlierRemoval(numeric_col.quantile(0.25), numeric_col.quantile(0.75))\n",
    "main_df[\"linksCount\"] = numeric_col.apply(outlier_remover_10.removeOutlier)\n",
    "\n",
    "sns.boxplot(x=main_df[\"linksCount\"])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "a6edcaf0",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWAAAAEGCAYAAABbzE8LAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAAALJUlEQVR4nO3dX6ykd13H8c+322K3Fq24TWnWyopLYgBNLSvUKFr/BnuDFxggpgg3BhOX5aJRwETFG4UYTV0vSI1NIDYqiYCELBoV+ieRVnebQlta4dgU7VKlUNlSd9vS+vNiZu3JyW53dj0z3/Psvl7JSefMPDPP77e/yfvMeebM0xpjBIDVO697AADnKgEGaCLAAE0EGKCJAAM0Of90Nt6xY8fYtWvXkoYCcHY6dOjQV8cYl268/rQCvGvXrhw8eHDzRgVwDqiqL53oeocgAJoIMEATAQZoIsAATQQYoIkAAzQRYIAmAgzQRIABmggwQBMBBmgiwABNBBigiQADNBFggCYCDNBEgAGaCDBAEwEGaHJa/084zm379+/P2tpa9zBOy+HDh5MkO3fubNn/7t27s3fv3pZ9s/UJMAtbW1vL3ffen2cvelH3UBa27eiRJMl/PLX6p/q2o4+tfJ9MiwBzWp696EU59n3Xdg9jYdsfOJAkLWM+vm84GceAAZoIMEATAQZoIsAATQQYoIkAAzQRYIAmAgzQRIABmggwQBMBBmgiwABNBBigiQADNBFggCYCDNBEgAGaCDBAEwEGaCLAAE0EGKCJAAM0EWCAJgIM0ESAAZoIMEATAQZoIsAATQQYoIkAAzQRYIAmAgzQRIABmggwQBMBBmgiwABNBBigiQADNBFggCYCDNBEgAGarCTA+/fvz/79+1exK4BNtcx+nb+UR91gbW1tFbsB2HTL7JdDEABNBBigiQADNBFggCYCDNBEgAGaCDBAEwEGaCLAAE0EGKCJAAM0EWCAJgIM0ESAAZoIMEATAQZoIsAATQQYoIkAAzQRYIAmAgzQRIABmggwQBMBBmgiwABNBBigiQADNBFggCYCDNBEgAGaCDBAEwEGaCLAAE0EGKCJAAM0EWCAJgIM0ESAAZoIMEATAQZoIsAATc5fxU4OHz6cY8eOZd++favYHUuytraW854e3cOYjPOefDxra9/wvJ+4tbW1bN++fSmPfcpXwFX1y1V1sKoOPvroo0sZBMC56JSvgMcYNya5MUn27NlzRi9/du7cmSS54YYbzuTubBH79u3LoQf/s3sYk/E/F35bdr/0Ms/7iVvmbzCOAQM0EWCAJgIM0ESAAZoIMEATAQZoIsAATQQYoIkAAzQRYIAmAgzQRIABmggwQBMBBmgiwABNBBigiQADNBFggCYCDNBEgAGaCDBAEwEGaCLAAE0EGKCJAAM0EWCAJgIM0ESAAZoIMEATAQZoIsAATQQYoIkAAzQRYIAmAgzQRIABmggwQBMBBmgiwABNBBigiQADNBFggCbnr2Inu3fvXsVuADbdMvu1kgDv3bt3FbsB2HTL7JdDEABNBBigiQADNBFggCYCDNBEgAGaCDBAEwEGaCLAAE0EGKCJAAM0EWCAJgIM0ESAAZoIMEATAQZoIsAATQQYoIkAAzQRYIAmAgzQRIABmggwQBMBBmgiwABNBBigiQADNBFggCYCDNBEgAGaCDBAEwEGaCLAAE0EGKCJAAM0EWCAJgIM0ESAAZoIMEATAQZocn73AJiWbUcfy/YHDnQPY2Hbjn4tSVrGvO3oY0kuW/l+mQ4BZmG7d+/uHsJpO3z4mSTJzp0dIbxskv9mrI4As7C9e/d2DwHOKo4BAzQRYIAmAgzQRIABmggwQBMBBmgiwABNBBigiQADNBFggCYCDNBEgAGaCDBAEwEGaCLAAE0EGKCJAAM0EWCAJgIM0ESAAZrUGGPxjaseTfKl5Q3nlHYk+Wrj/pfJ3KbnbJ1XYm6b7SVjjEs3XnlaAe5WVQfHGHu6x7EM5jY9Z+u8EnNbFYcgAJoIMECTqQX4xu4BLJG5Tc/ZOq/E3FZiUseAAc4mU3sFDHDWEGCAJlsuwFX1uqr6l6paq6p3neD2a6rqSFXdPf/6zY5xnomquqmqvlJV957k9qqqP5rP/XNVddWqx3imFpjbJNetqq6oqk9X1f1VdV9V7TvBNpNctwXnNtV1u7Cq/qmqPjuf23tPsE3/uo0xtsxXkm1J/jXJS5O8IMlnk7x8wzbXJPlE91jPcH4/luSqJPee5PZrk3wySSW5Osmd3WPexLlNct2SXJ7kqvnlFyb5wgmek5NctwXnNtV1qyQXzy9fkOTOJFdvtXXbaq+AX51kbYzx4Bjj6SR/keT1zWPaNGOM25I89jybvD7Jh8bMHUkuqarLVzO6/58F5jZJY4xHxhh3zS9/I8n9SXZu2GyS67bg3CZpvhZPzL+9YP618S8O2tdtqwV4Z5J/X/f9wznxE+KH579afLKqXrGaoa3EovOfqkmvW1XtSvKDmb2aWm/y6/Y8c0smum5Vta2q7k7ylSR/N8bYcut2/ip3toA6wXUbf2rdldnnqp+oqmuTfCzJy5Y9sBVZZP5TNel1q6qLk/xVkneOMR7fePMJ7jKZdTvF3Ca7bmOMZ5NcWVWXJPloVb1yjLH+PYr2ddtqr4AfTnLFuu+/K8mX128wxnj8+K8WY4wDSS6oqh2rG+JSnXL+UzXldauqCzIL1M1jjI+cYJPJrtup5jbldTtujPH1JLcked2Gm9rXbasF+J+TvKyqvqeqXpDkTUk+vn6DqnpxVdX88qszm8PXVj7S5fh4krfM3529OsmRMcYj3YPaDFNdt/mY/zTJ/WOMPzjJZpNct0XmNuF1u3T+yjdVtT3JTyd5YMNm7eu2pQ5BjDGeqapfTfK3mf1FxE1jjPuq6u3z2z+Q5A1JfqWqnklyLMmbxvwtza2uqv48s3eVd1TVw0l+K7M3B47P7UBm78yuJTma5G09Iz19C8xtquv2I0muS3LP/HhikrwnyXcnk1+3ReY21XW7PMkHq2pbZj80PjzG+MSGlrSvm48iAzTZaocgAM4ZAgzQRIABmggwQBMBBmgiwExOVb21qv543fdvqap752e9+nxVXb+Efb5nsx8TBJgtb/63nCe77eeSvDPJz44xXpHZGdmOLGEYAsymE2CWqqp+rareMb/8h1X1qfnln6qqP6uqN1fVPfNXsO9bd78nqup3qurOzE4G87aq+kJV3ZrZBwiOe3eS68cYX06SMcaTY4w/mT/GlVV1x/xcrx+tqu+YX39LVe2ZX95RVQ/NL7+1qj5SVX9TVV+sqvfPr/+9JNtrdj7cm5f7L8a5RIBZttuSvHZ+eU+Si+fnH/jRJF9M8r4kP5nkyiQ/VFU/P9/2WzM7t/BrMjtH9HszC+/PJHn5usd/ZZJDJ9n3h5L8+hjjB5Lck9mn807lyiRvTPL9Sd5YVVeMMd6V5NgY48oxxi8u8BiwEAFm2Q4leVVVvTDJU0k+k1mIX5vk60luGWM8OsZ4JsnNmZ3YPUmezewkMUnymnXbPZ3kL0+106r69iSXjDFunV/1wXWP/Xz+YYxxZIzxZJLPJ3nJAveBMyLALNUY45tJHsrsc/b/mOT2JD+R5HuT/Nvz3PXJ+ekE/++hTrLdfUledZrDeibPPfcv3HDbU+suP5stdr4Uzi4CzCrcluT6+X9vT/L2JHcnuSPJj8+Pw25L8uYkt57g/ncmuaaqvnN++OIX1t32u0neX1UvTpKq+paqescY40iS/6qq44c/rlv32A/luWi/YcE5fHO+b9g0frqzCrcn+Y0knxlj/HdVPZnk9jHGI1X17iSfzuzk2AfGGH+98c7z7X47s8MXj2R2kvBt89sOVNVlSf5+ftrEkeSm+V1/KckHquqiJA/mubNd/X6SD1fVdUk+teAcbkzyuaq6y3FgNouzoQE0cQgCoIkAAzQRYIAmAgzQRIABmggwQBMBBmjyvz/FX/ZW8Z6OAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "numeric_col = main_df[\"wordCount\"]\n",
    "sns.boxplot(x=numeric_col)\n",
    "plt.show()\n",
    "\n",
    "outlier_remover_11 = OutlierRemoval(numeric_col.quantile(0.25), numeric_col.quantile(0.75))\n",
    "main_df[\"wordCount\"] = numeric_col.apply(outlier_remover_11.removeOutlier)\n",
    "\n",
    "sns.boxplot(x=main_df[\"wordCount\"])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "1fefa98b",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "numeric_col = main_df[\"parametrizedLinkRatio\"]\n",
    "sns.boxplot(x=numeric_col)\n",
    "plt.show()\n",
    "\n",
    "outlier_remover_12 = OutlierRemoval(numeric_col.quantile(0.25), numeric_col.quantile(0.75))\n",
    "main_df[\"parametrizedLinkRatio\"] = numeric_col.apply(outlier_remover_12.removeOutlier)\n",
    "\n",
    "sns.boxplot(x=main_df[\"parametrizedLinkRatio\"])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "69995fca",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "numeric_col = main_df[\"spellingErrorsRatio\"]\n",
    "sns.boxplot(x=numeric_col)\n",
    "plt.show()\n",
    "\n",
    "outlier_remover_13 = OutlierRemoval(numeric_col.quantile(0.25), numeric_col.quantile(0.75))\n",
    "main_df[\"spellingErrorsRatio\"] = numeric_col.apply(outlier_remover_13.removeOutlier)\n",
    "\n",
    "sns.boxplot(x=main_df[\"spellingErrorsRatio\"])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c498b407",
   "metadata": {},
   "source": [
    "## 1.4 One-hot encoding"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "09dd6ca8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "unknown               2348\n",
       "recreation            1229\n",
       "arts_entertainment     941\n",
       "business               880\n",
       "health                 506\n",
       "sports                 380\n",
       "culture_politics       343\n",
       "computer_internet      296\n",
       "science_technology     289\n",
       "gaming                  76\n",
       "religion                72\n",
       "law_crime               31\n",
       "weather                  4\n",
       "Name: alchemy_category, dtype: int64"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df[\"alchemy_category\"].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0ae14c2f",
   "metadata": {},
   "source": [
    "Alchemy category is one-hot encoded"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "44d1e14c",
   "metadata": {},
   "outputs": [],
   "source": [
    "column_names_to_one_hot = [\"alchemy_category\"]\n",
    "main_df = pd.get_dummies(main_df, columns=column_names_to_one_hot)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "6b6b93cb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>url</th>\n",
       "      <th>webpageDescription</th>\n",
       "      <th>alchemy_category_score</th>\n",
       "      <th>avgLinkWordLength</th>\n",
       "      <th>AvglinkWithOneCommonWord</th>\n",
       "      <th>AvglinkWithTwoCommonWord</th>\n",
       "      <th>AvglinkWithThreeCommonWord</th>\n",
       "      <th>AvglinkWithFourCommonWord</th>\n",
       "      <th>redundancyMeasure</th>\n",
       "      <th>frameTagRatio</th>\n",
       "      <th>...</th>\n",
       "      <th>alchemy_category_culture_politics</th>\n",
       "      <th>alchemy_category_gaming</th>\n",
       "      <th>alchemy_category_health</th>\n",
       "      <th>alchemy_category_law_crime</th>\n",
       "      <th>alchemy_category_recreation</th>\n",
       "      <th>alchemy_category_religion</th>\n",
       "      <th>alchemy_category_science_technology</th>\n",
       "      <th>alchemy_category_sports</th>\n",
       "      <th>alchemy_category_unknown</th>\n",
       "      <th>alchemy_category_weather</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>http://www.polyvore.com/cgi/home?id=1389651</td>\n",
       "      <td>{\"title\":\"Join Polyvore and follow the latest ...</td>\n",
       "      <td>1.869397</td>\n",
       "      <td>1.070441</td>\n",
       "      <td>2.850855</td>\n",
       "      <td>0.007905</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.589894</td>\n",
       "      <td>0.027778</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>http://www.youtube.com/watch?v=ippMPPu6gh4</td>\n",
       "      <td>{\"body\":\"Speed Air Man--David Belle david bell...</td>\n",
       "      <td>1.869397</td>\n",
       "      <td>0.814292</td>\n",
       "      <td>3.129977</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.591068</td>\n",
       "      <td>0.015086</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>http://www.musingsofahousewife.com/2011/03/tri...</td>\n",
       "      <td>{\"title\":\"Chicken Gruyere Recipe \",\"body\":\"Chi...</td>\n",
       "      <td>1.472093</td>\n",
       "      <td>1.106580</td>\n",
       "      <td>5.105662</td>\n",
       "      <td>0.250017</td>\n",
       "      <td>0.209826</td>\n",
       "      <td>0.184265</td>\n",
       "      <td>0.367008</td>\n",
       "      <td>0.033935</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>http://www.thelittleteochew.com/2011/07/ikan-b...</td>\n",
       "      <td>{\"title\":\"Ikan Bilis With Tempeh Groundnuts \",...</td>\n",
       "      <td>1.608077</td>\n",
       "      <td>0.980299</td>\n",
       "      <td>4.135385</td>\n",
       "      <td>0.181739</td>\n",
       "      <td>0.093401</td>\n",
       "      <td>0.064320</td>\n",
       "      <td>0.387063</td>\n",
       "      <td>0.036530</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>http://recipes.wuzzle.org/index.php/72</td>\n",
       "      <td>{\"url\":\"recipes wuzzle org index php 72\",\"titl...</td>\n",
       "      <td>1.707464</td>\n",
       "      <td>0.457966</td>\n",
       "      <td>2.818948</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.256666</td>\n",
       "      <td>0.015152</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 36 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 url  \\\n",
       "0        http://www.polyvore.com/cgi/home?id=1389651   \n",
       "1         http://www.youtube.com/watch?v=ippMPPu6gh4   \n",
       "2  http://www.musingsofahousewife.com/2011/03/tri...   \n",
       "3  http://www.thelittleteochew.com/2011/07/ikan-b...   \n",
       "4             http://recipes.wuzzle.org/index.php/72   \n",
       "\n",
       "                                  webpageDescription  alchemy_category_score  \\\n",
       "0  {\"title\":\"Join Polyvore and follow the latest ...                1.869397   \n",
       "1  {\"body\":\"Speed Air Man--David Belle david bell...                1.869397   \n",
       "2  {\"title\":\"Chicken Gruyere Recipe \",\"body\":\"Chi...                1.472093   \n",
       "3  {\"title\":\"Ikan Bilis With Tempeh Groundnuts \",...                1.608077   \n",
       "4  {\"url\":\"recipes wuzzle org index php 72\",\"titl...                1.707464   \n",
       "\n",
       "   avgLinkWordLength  AvglinkWithOneCommonWord  AvglinkWithTwoCommonWord  \\\n",
       "0           1.070441                  2.850855                  0.007905   \n",
       "1           0.814292                  3.129977                  0.000000   \n",
       "2           1.106580                  5.105662                  0.250017   \n",
       "3           0.980299                  4.135385                  0.181739   \n",
       "4           0.457966                  2.818948                  0.000000   \n",
       "\n",
       "   AvglinkWithThreeCommonWord  AvglinkWithFourCommonWord  redundancyMeasure  \\\n",
       "0                    0.000000                   0.000000           0.589894   \n",
       "1                    0.000000                   0.000000           0.591068   \n",
       "2                    0.209826                   0.184265           0.367008   \n",
       "3                    0.093401                   0.064320           0.387063   \n",
       "4                    0.000000                   0.000000           0.256666   \n",
       "\n",
       "   frameTagRatio  ...  alchemy_category_culture_politics  \\\n",
       "0       0.027778  ...                                  0   \n",
       "1       0.015086  ...                                  0   \n",
       "2       0.033935  ...                                  0   \n",
       "3       0.036530  ...                                  0   \n",
       "4       0.015152  ...                                  0   \n",
       "\n",
       "   alchemy_category_gaming alchemy_category_health  \\\n",
       "0                        0                       0   \n",
       "1                        0                       0   \n",
       "2                        0                       0   \n",
       "3                        0                       0   \n",
       "4                        0                       0   \n",
       "\n",
       "   alchemy_category_law_crime  alchemy_category_recreation  \\\n",
       "0                           0                            0   \n",
       "1                           0                            0   \n",
       "2                           0                            0   \n",
       "3                           0                            1   \n",
       "4                           0                            0   \n",
       "\n",
       "   alchemy_category_religion  alchemy_category_science_technology  \\\n",
       "0                          0                                    0   \n",
       "1                          0                                    0   \n",
       "2                          0                                    1   \n",
       "3                          0                                    0   \n",
       "4                          0                                    0   \n",
       "\n",
       "   alchemy_category_sports  alchemy_category_unknown  alchemy_category_weather  \n",
       "0                        0                         1                         0  \n",
       "1                        0                         1                         0  \n",
       "2                        0                         0                         0  \n",
       "3                        0                         0                         0  \n",
       "4                        0                         0                         0  \n",
       "\n",
       "[5 rows x 36 columns]"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f8629242",
   "metadata": {},
   "source": [
    "## 1.5 Replacing webpageDescription with 2-columns \"body\", \"title\"."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "9970e2ac",
   "metadata": {},
   "outputs": [],
   "source": [
    "col = main_df[\"webpageDescription\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "8b533d24",
   "metadata": {},
   "outputs": [],
   "source": [
    "col_name = \"body\"\n",
    "col = col.to_numpy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "8737f0af",
   "metadata": {},
   "outputs": [],
   "source": [
    "col_list = []\n",
    "for i in range(len(col)):\n",
    "    a = json.loads(col[i])\n",
    "    if \"body\" in a.keys() and a[\"body\"]!=' ':\n",
    "        col_list.append(a[\"body\"])\n",
    "    else:\n",
    "        col_list.append(\"Unkonown\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "8e73ba99",
   "metadata": {},
   "outputs": [],
   "source": [
    "main_df.insert(1, col_name, col_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "6f8b24e4",
   "metadata": {},
   "outputs": [],
   "source": [
    "col_name = \"title\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "523d47d7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "7395"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "col_list_title = []\n",
    "for i in range(len(col)):\n",
    "    a = json.loads(col[i])\n",
    "    if \"title\" in a.keys() and a[\"title\"]!=' ':\n",
    "        col_list_title.append(a[\"title\"])\n",
    "    else:\n",
    "        col_list_title.append(\"Unkonown\")\n",
    "        \n",
    "len(col_list_title)  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "591ece4d",
   "metadata": {},
   "outputs": [],
   "source": [
    "main_df.insert(1, col_name, col_list_title)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "5a1a2a98",
   "metadata": {},
   "outputs": [],
   "source": [
    "main_df.drop(axis=\"columns\", labels=\"webpageDescription\", inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "b4fb3851",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>url</th>\n",
       "      <th>title</th>\n",
       "      <th>body</th>\n",
       "      <th>alchemy_category_score</th>\n",
       "      <th>avgLinkWordLength</th>\n",
       "      <th>AvglinkWithOneCommonWord</th>\n",
       "      <th>AvglinkWithTwoCommonWord</th>\n",
       "      <th>AvglinkWithThreeCommonWord</th>\n",
       "      <th>AvglinkWithFourCommonWord</th>\n",
       "      <th>redundancyMeasure</th>\n",
       "      <th>...</th>\n",
       "      <th>alchemy_category_culture_politics</th>\n",
       "      <th>alchemy_category_gaming</th>\n",
       "      <th>alchemy_category_health</th>\n",
       "      <th>alchemy_category_law_crime</th>\n",
       "      <th>alchemy_category_recreation</th>\n",
       "      <th>alchemy_category_religion</th>\n",
       "      <th>alchemy_category_science_technology</th>\n",
       "      <th>alchemy_category_sports</th>\n",
       "      <th>alchemy_category_unknown</th>\n",
       "      <th>alchemy_category_weather</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>http://www.polyvore.com/cgi/home?id=1389651</td>\n",
       "      <td>Join Polyvore and follow the latest styles</td>\n",
       "      <td>polyvore is the best place to discover or sta...</td>\n",
       "      <td>1.869397</td>\n",
       "      <td>1.070441</td>\n",
       "      <td>2.850855</td>\n",
       "      <td>0.007905</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.589894</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>http://www.youtube.com/watch?v=ippMPPu6gh4</td>\n",
       "      <td>Speed Air Man--David Belle</td>\n",
       "      <td>Speed Air Man--David Belle david belle speed a...</td>\n",
       "      <td>1.869397</td>\n",
       "      <td>0.814292</td>\n",
       "      <td>3.129977</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.591068</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>http://www.musingsofahousewife.com/2011/03/tri...</td>\n",
       "      <td>Chicken Gruyere Recipe</td>\n",
       "      <td>Chicken Gruyere one of our favorite special di...</td>\n",
       "      <td>1.472093</td>\n",
       "      <td>1.106580</td>\n",
       "      <td>5.105662</td>\n",
       "      <td>0.250017</td>\n",
       "      <td>0.209826</td>\n",
       "      <td>0.184265</td>\n",
       "      <td>0.367008</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>http://www.thelittleteochew.com/2011/07/ikan-b...</td>\n",
       "      <td>Ikan Bilis With Tempeh Groundnuts</td>\n",
       "      <td>Oh me oh my This was really snackalicious swee...</td>\n",
       "      <td>1.608077</td>\n",
       "      <td>0.980299</td>\n",
       "      <td>4.135385</td>\n",
       "      <td>0.181739</td>\n",
       "      <td>0.093401</td>\n",
       "      <td>0.064320</td>\n",
       "      <td>0.387063</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>http://recipes.wuzzle.org/index.php/72</td>\n",
       "      <td>Malaysian recipes</td>\n",
       "      <td>Barbecued Chicken Chow Siew from The Exotic Ki...</td>\n",
       "      <td>1.707464</td>\n",
       "      <td>0.457966</td>\n",
       "      <td>2.818948</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.256666</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 37 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 url  \\\n",
       "0        http://www.polyvore.com/cgi/home?id=1389651   \n",
       "1         http://www.youtube.com/watch?v=ippMPPu6gh4   \n",
       "2  http://www.musingsofahousewife.com/2011/03/tri...   \n",
       "3  http://www.thelittleteochew.com/2011/07/ikan-b...   \n",
       "4             http://recipes.wuzzle.org/index.php/72   \n",
       "\n",
       "                                         title  \\\n",
       "0  Join Polyvore and follow the latest styles    \n",
       "1                   Speed Air Man--David Belle   \n",
       "2                      Chicken Gruyere Recipe    \n",
       "3           Ikan Bilis With Tempeh Groundnuts    \n",
       "4                           Malaysian recipes    \n",
       "\n",
       "                                                body  alchemy_category_score  \\\n",
       "0   polyvore is the best place to discover or sta...                1.869397   \n",
       "1  Speed Air Man--David Belle david belle speed a...                1.869397   \n",
       "2  Chicken Gruyere one of our favorite special di...                1.472093   \n",
       "3  Oh me oh my This was really snackalicious swee...                1.608077   \n",
       "4  Barbecued Chicken Chow Siew from The Exotic Ki...                1.707464   \n",
       "\n",
       "   avgLinkWordLength  AvglinkWithOneCommonWord  AvglinkWithTwoCommonWord  \\\n",
       "0           1.070441                  2.850855                  0.007905   \n",
       "1           0.814292                  3.129977                  0.000000   \n",
       "2           1.106580                  5.105662                  0.250017   \n",
       "3           0.980299                  4.135385                  0.181739   \n",
       "4           0.457966                  2.818948                  0.000000   \n",
       "\n",
       "   AvglinkWithThreeCommonWord  AvglinkWithFourCommonWord  redundancyMeasure  \\\n",
       "0                    0.000000                   0.000000           0.589894   \n",
       "1                    0.000000                   0.000000           0.591068   \n",
       "2                    0.209826                   0.184265           0.367008   \n",
       "3                    0.093401                   0.064320           0.387063   \n",
       "4                    0.000000                   0.000000           0.256666   \n",
       "\n",
       "   ...  alchemy_category_culture_politics  alchemy_category_gaming  \\\n",
       "0  ...                                  0                        0   \n",
       "1  ...                                  0                        0   \n",
       "2  ...                                  0                        0   \n",
       "3  ...                                  0                        0   \n",
       "4  ...                                  0                        0   \n",
       "\n",
       "   alchemy_category_health alchemy_category_law_crime  \\\n",
       "0                        0                          0   \n",
       "1                        0                          0   \n",
       "2                        0                          0   \n",
       "3                        0                          0   \n",
       "4                        0                          0   \n",
       "\n",
       "   alchemy_category_recreation  alchemy_category_religion  \\\n",
       "0                            0                          0   \n",
       "1                            0                          0   \n",
       "2                            0                          0   \n",
       "3                            1                          0   \n",
       "4                            0                          0   \n",
       "\n",
       "   alchemy_category_science_technology  alchemy_category_sports  \\\n",
       "0                                    0                        0   \n",
       "1                                    0                        0   \n",
       "2                                    1                        0   \n",
       "3                                    0                        0   \n",
       "4                                    0                        0   \n",
       "\n",
       "   alchemy_category_unknown  alchemy_category_weather  \n",
       "0                         1                         0  \n",
       "1                         1                         0  \n",
       "2                         0                         0  \n",
       "3                         0                         0  \n",
       "4                         0                         0  \n",
       "\n",
       "[5 rows x 37 columns]"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d5055b5d",
   "metadata": {},
   "source": [
    "## 1.6 Modifying url column by deleting symbols and strings which are not useful in predictions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "da1b312a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['http://www.polyvore.com/cgi/home?id=1389651',\n",
       "       'http://www.youtube.com/watch?v=ippMPPu6gh4',\n",
       "       'http://www.musingsofahousewife.com/2011/03/tried-and-true-chicken-gruyere.html',\n",
       "       ...,\n",
       "       'http://www.laweekly.com/bestof/2011/award/best-belly-dance-workout-1509632/',\n",
       "       'http://www.kraftrecipes.com/recipes/buffalo-chicken-burgers-118650.aspx?cm_mmc=eml-_-rbe-_-20110526-_-1020',\n",
       "       'http://canarygirl.com/?p=80'], dtype=object)"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "col = main_df[\"url\"]\n",
    "col = col.to_numpy()\n",
    "col"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "549b7118",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['polyvore cgi home id 1389651', 'youtube watch v ippMPPu6gh4',\n",
       "       'musingsofahousewife 2011 03 tried and true chicken gruyere html',\n",
       "       ..., 'laweekly bestof 2011 award best belly dance workout 1509632',\n",
       "       'kraftrecipes recipes buffalo chicken burgers 118650 aspx cm mmc eml rbe 20110526 1020',\n",
       "       'canarygirl p 80'], dtype=object)"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "charlist = ['/','?','=','-','_','&','(',')','@','$','%','#',':',';',\"<\",\">\",\".\"]\n",
    "for i in range(len(col)):\n",
    "    col[i] = col[i].replace('http://', '')\n",
    "    col[i] = col[i].replace('https://', '')\n",
    "    col[i] = col[i].replace('www.','').replace('.com','').replace('.org','')\n",
    "    for j in range(len(charlist)):\n",
    "        col[i] = col[i].replace(charlist[j],' ')\n",
    "    col[i] = \" \".join(col[i].split()) \n",
    "    \n",
    "    \n",
    "col"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "4ee7928d",
   "metadata": {},
   "outputs": [],
   "source": [
    "main_df[\"url\"] = col"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "89816e05",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>url</th>\n",
       "      <th>title</th>\n",
       "      <th>body</th>\n",
       "      <th>alchemy_category_score</th>\n",
       "      <th>avgLinkWordLength</th>\n",
       "      <th>AvglinkWithOneCommonWord</th>\n",
       "      <th>AvglinkWithTwoCommonWord</th>\n",
       "      <th>AvglinkWithThreeCommonWord</th>\n",
       "      <th>AvglinkWithFourCommonWord</th>\n",
       "      <th>redundancyMeasure</th>\n",
       "      <th>...</th>\n",
       "      <th>alchemy_category_culture_politics</th>\n",
       "      <th>alchemy_category_gaming</th>\n",
       "      <th>alchemy_category_health</th>\n",
       "      <th>alchemy_category_law_crime</th>\n",
       "      <th>alchemy_category_recreation</th>\n",
       "      <th>alchemy_category_religion</th>\n",
       "      <th>alchemy_category_science_technology</th>\n",
       "      <th>alchemy_category_sports</th>\n",
       "      <th>alchemy_category_unknown</th>\n",
       "      <th>alchemy_category_weather</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>polyvore cgi home id 1389651</td>\n",
       "      <td>Join Polyvore and follow the latest styles</td>\n",
       "      <td>polyvore is the best place to discover or sta...</td>\n",
       "      <td>1.869397</td>\n",
       "      <td>1.070441</td>\n",
       "      <td>2.850855</td>\n",
       "      <td>0.007905</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.589894</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>youtube watch v ippMPPu6gh4</td>\n",
       "      <td>Speed Air Man--David Belle</td>\n",
       "      <td>Speed Air Man--David Belle david belle speed a...</td>\n",
       "      <td>1.869397</td>\n",
       "      <td>0.814292</td>\n",
       "      <td>3.129977</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.591068</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>musingsofahousewife 2011 03 tried and true chi...</td>\n",
       "      <td>Chicken Gruyere Recipe</td>\n",
       "      <td>Chicken Gruyere one of our favorite special di...</td>\n",
       "      <td>1.472093</td>\n",
       "      <td>1.106580</td>\n",
       "      <td>5.105662</td>\n",
       "      <td>0.250017</td>\n",
       "      <td>0.209826</td>\n",
       "      <td>0.184265</td>\n",
       "      <td>0.367008</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>thelittleteochew 2011 07 ikan bilis with tempe...</td>\n",
       "      <td>Ikan Bilis With Tempeh Groundnuts</td>\n",
       "      <td>Oh me oh my This was really snackalicious swee...</td>\n",
       "      <td>1.608077</td>\n",
       "      <td>0.980299</td>\n",
       "      <td>4.135385</td>\n",
       "      <td>0.181739</td>\n",
       "      <td>0.093401</td>\n",
       "      <td>0.064320</td>\n",
       "      <td>0.387063</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>recipes wuzzle index php 72</td>\n",
       "      <td>Malaysian recipes</td>\n",
       "      <td>Barbecued Chicken Chow Siew from The Exotic Ki...</td>\n",
       "      <td>1.707464</td>\n",
       "      <td>0.457966</td>\n",
       "      <td>2.818948</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.256666</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1474</th>\n",
       "      <td>lanascooking 2011 07 08 roasted potato salad</td>\n",
       "      <td>Unkonown</td>\n",
       "      <td>Unkonown</td>\n",
       "      <td>1.976679</td>\n",
       "      <td>0.970779</td>\n",
       "      <td>3.925608</td>\n",
       "      <td>0.097638</td>\n",
       "      <td>0.033617</td>\n",
       "      <td>0.016950</td>\n",
       "      <td>0.591068</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1475</th>\n",
       "      <td>marthastewart 255456 billys chocolate buttercream</td>\n",
       "      <td>Billy s Chocolate Buttercream Martha Stewart R...</td>\n",
       "      <td>Save to your Collections Sorry for the inconve...</td>\n",
       "      <td>1.869397</td>\n",
       "      <td>1.174511</td>\n",
       "      <td>5.348241</td>\n",
       "      <td>0.208901</td>\n",
       "      <td>0.087011</td>\n",
       "      <td>0.020001</td>\n",
       "      <td>0.345951</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1476</th>\n",
       "      <td>laweekly bestof 2011 award best belly dance wo...</td>\n",
       "      <td>Los Angeles Best Belly Dance Workout Swerve St...</td>\n",
       "      <td>Best Belly Dance Workout 2011 Swerve Studio s ...</td>\n",
       "      <td>2.505420</td>\n",
       "      <td>1.040202</td>\n",
       "      <td>4.538926</td>\n",
       "      <td>0.246225</td>\n",
       "      <td>0.054334</td>\n",
       "      <td>0.020101</td>\n",
       "      <td>0.436514</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1477</th>\n",
       "      <td>kraftrecipes recipes buffalo chicken burgers 1...</td>\n",
       "      <td>Buffalo Chicken Burgers recipe buffalo chicken...</td>\n",
       "      <td>outerbanxchic posted 10 14 2011 Delicious and ...</td>\n",
       "      <td>1.869397</td>\n",
       "      <td>1.009167</td>\n",
       "      <td>4.754497</td>\n",
       "      <td>0.204794</td>\n",
       "      <td>0.078643</td>\n",
       "      <td>0.018019</td>\n",
       "      <td>0.453249</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1478</th>\n",
       "      <td>canarygirl p 80</td>\n",
       "      <td>canarygirl com Sirloin with Peppercorn Cream S...</td>\n",
       "      <td>Summer always signifies the beginning of steak...</td>\n",
       "      <td>2.044258</td>\n",
       "      <td>1.105337</td>\n",
       "      <td>4.445692</td>\n",
       "      <td>0.196998</td>\n",
       "      <td>0.031749</td>\n",
       "      <td>0.027835</td>\n",
       "      <td>0.409636</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>7395 rows × 37 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    url  \\\n",
       "0                          polyvore cgi home id 1389651   \n",
       "1                           youtube watch v ippMPPu6gh4   \n",
       "2     musingsofahousewife 2011 03 tried and true chi...   \n",
       "3     thelittleteochew 2011 07 ikan bilis with tempe...   \n",
       "4                           recipes wuzzle index php 72   \n",
       "...                                                 ...   \n",
       "1474       lanascooking 2011 07 08 roasted potato salad   \n",
       "1475  marthastewart 255456 billys chocolate buttercream   \n",
       "1476  laweekly bestof 2011 award best belly dance wo...   \n",
       "1477  kraftrecipes recipes buffalo chicken burgers 1...   \n",
       "1478                                    canarygirl p 80   \n",
       "\n",
       "                                                  title  \\\n",
       "0           Join Polyvore and follow the latest styles    \n",
       "1                            Speed Air Man--David Belle   \n",
       "2                               Chicken Gruyere Recipe    \n",
       "3                    Ikan Bilis With Tempeh Groundnuts    \n",
       "4                                    Malaysian recipes    \n",
       "...                                                 ...   \n",
       "1474                                           Unkonown   \n",
       "1475  Billy s Chocolate Buttercream Martha Stewart R...   \n",
       "1476  Los Angeles Best Belly Dance Workout Swerve St...   \n",
       "1477  Buffalo Chicken Burgers recipe buffalo chicken...   \n",
       "1478  canarygirl com Sirloin with Peppercorn Cream S...   \n",
       "\n",
       "                                                   body  \\\n",
       "0      polyvore is the best place to discover or sta...   \n",
       "1     Speed Air Man--David Belle david belle speed a...   \n",
       "2     Chicken Gruyere one of our favorite special di...   \n",
       "3     Oh me oh my This was really snackalicious swee...   \n",
       "4     Barbecued Chicken Chow Siew from The Exotic Ki...   \n",
       "...                                                 ...   \n",
       "1474                                           Unkonown   \n",
       "1475  Save to your Collections Sorry for the inconve...   \n",
       "1476  Best Belly Dance Workout 2011 Swerve Studio s ...   \n",
       "1477  outerbanxchic posted 10 14 2011 Delicious and ...   \n",
       "1478  Summer always signifies the beginning of steak...   \n",
       "\n",
       "      alchemy_category_score  avgLinkWordLength  AvglinkWithOneCommonWord  \\\n",
       "0                   1.869397           1.070441                  2.850855   \n",
       "1                   1.869397           0.814292                  3.129977   \n",
       "2                   1.472093           1.106580                  5.105662   \n",
       "3                   1.608077           0.980299                  4.135385   \n",
       "4                   1.707464           0.457966                  2.818948   \n",
       "...                      ...                ...                       ...   \n",
       "1474                1.976679           0.970779                  3.925608   \n",
       "1475                1.869397           1.174511                  5.348241   \n",
       "1476                2.505420           1.040202                  4.538926   \n",
       "1477                1.869397           1.009167                  4.754497   \n",
       "1478                2.044258           1.105337                  4.445692   \n",
       "\n",
       "      AvglinkWithTwoCommonWord  AvglinkWithThreeCommonWord  \\\n",
       "0                     0.007905                    0.000000   \n",
       "1                     0.000000                    0.000000   \n",
       "2                     0.250017                    0.209826   \n",
       "3                     0.181739                    0.093401   \n",
       "4                     0.000000                    0.000000   \n",
       "...                        ...                         ...   \n",
       "1474                  0.097638                    0.033617   \n",
       "1475                  0.208901                    0.087011   \n",
       "1476                  0.246225                    0.054334   \n",
       "1477                  0.204794                    0.078643   \n",
       "1478                  0.196998                    0.031749   \n",
       "\n",
       "      AvglinkWithFourCommonWord  redundancyMeasure  ...  \\\n",
       "0                      0.000000           0.589894  ...   \n",
       "1                      0.000000           0.591068  ...   \n",
       "2                      0.184265           0.367008  ...   \n",
       "3                      0.064320           0.387063  ...   \n",
       "4                      0.000000           0.256666  ...   \n",
       "...                         ...                ...  ...   \n",
       "1474                   0.016950           0.591068  ...   \n",
       "1475                   0.020001           0.345951  ...   \n",
       "1476                   0.020101           0.436514  ...   \n",
       "1477                   0.018019           0.453249  ...   \n",
       "1478                   0.027835           0.409636  ...   \n",
       "\n",
       "      alchemy_category_culture_politics  alchemy_category_gaming  \\\n",
       "0                                     0                        0   \n",
       "1                                     0                        0   \n",
       "2                                     0                        0   \n",
       "3                                     0                        0   \n",
       "4                                     0                        0   \n",
       "...                                 ...                      ...   \n",
       "1474                                  0                        0   \n",
       "1475                                  0                        0   \n",
       "1476                                  0                        0   \n",
       "1477                                  0                        0   \n",
       "1478                                  0                        0   \n",
       "\n",
       "      alchemy_category_health alchemy_category_law_crime  \\\n",
       "0                           0                          0   \n",
       "1                           0                          0   \n",
       "2                           0                          0   \n",
       "3                           0                          0   \n",
       "4                           0                          0   \n",
       "...                       ...                        ...   \n",
       "1474                        0                          0   \n",
       "1475                        0                          0   \n",
       "1476                        0                          0   \n",
       "1477                        0                          0   \n",
       "1478                        0                          0   \n",
       "\n",
       "      alchemy_category_recreation  alchemy_category_religion  \\\n",
       "0                               0                          0   \n",
       "1                               0                          0   \n",
       "2                               0                          0   \n",
       "3                               1                          0   \n",
       "4                               0                          0   \n",
       "...                           ...                        ...   \n",
       "1474                            0                          0   \n",
       "1475                            0                          0   \n",
       "1476                            0                          0   \n",
       "1477                            0                          0   \n",
       "1478                            0                          0   \n",
       "\n",
       "      alchemy_category_science_technology  alchemy_category_sports  \\\n",
       "0                                       0                        0   \n",
       "1                                       0                        0   \n",
       "2                                       1                        0   \n",
       "3                                       0                        0   \n",
       "4                                       0                        0   \n",
       "...                                   ...                      ...   \n",
       "1474                                    0                        0   \n",
       "1475                                    0                        0   \n",
       "1476                                    0                        0   \n",
       "1477                                    0                        0   \n",
       "1478                                    0                        0   \n",
       "\n",
       "      alchemy_category_unknown  alchemy_category_weather  \n",
       "0                            1                         0  \n",
       "1                            1                         0  \n",
       "2                            0                         0  \n",
       "3                            0                         0  \n",
       "4                            0                         0  \n",
       "...                        ...                       ...  \n",
       "1474                         0                         0  \n",
       "1475                         1                         0  \n",
       "1476                         0                         0  \n",
       "1477                         1                         0  \n",
       "1478                         0                         0  \n",
       "\n",
       "[7395 rows x 37 columns]"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "22311081",
   "metadata": {},
   "source": [
    "## 1.7 Removing null values in title and body columns."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "4d15a731",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "url                                       0\n",
       "title                                    12\n",
       "body                                     57\n",
       "alchemy_category_score                    0\n",
       "avgLinkWordLength                         0\n",
       "AvglinkWithOneCommonWord                  0\n",
       "AvglinkWithTwoCommonWord                  0\n",
       "AvglinkWithThreeCommonWord                0\n",
       "AvglinkWithFourCommonWord                 0\n",
       "redundancyMeasure                         0\n",
       "frameTagRatio                             0\n",
       "tagRatio                                  0\n",
       "imageTagRatio                             0\n",
       "isNews                                    0\n",
       "lengthyDomain                             0\n",
       "hyperlinkToAllWordsRatio                  0\n",
       "isFrontPageNews                           0\n",
       "alphanumCharCount                         0\n",
       "linksCount                                0\n",
       "wordCount                                 0\n",
       "parametrizedLinkRatio                     0\n",
       "spellingErrorsRatio                       0\n",
       "label                                  1479\n",
       "id                                        0\n",
       "alchemy_category_arts_entertainment       0\n",
       "alchemy_category_business                 0\n",
       "alchemy_category_computer_internet        0\n",
       "alchemy_category_culture_politics         0\n",
       "alchemy_category_gaming                   0\n",
       "alchemy_category_health                   0\n",
       "alchemy_category_law_crime                0\n",
       "alchemy_category_recreation               0\n",
       "alchemy_category_religion                 0\n",
       "alchemy_category_science_technology       0\n",
       "alchemy_category_sports                   0\n",
       "alchemy_category_unknown                  0\n",
       "alchemy_category_weather                  0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "c9518c66",
   "metadata": {},
   "outputs": [],
   "source": [
    "main_df[\"title\"].fillna(\"Unknown\", inplace=True)\n",
    "main_df[\"body\"].fillna(\"Unknown\", inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "e40bdc05",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "url                                       0\n",
       "title                                     0\n",
       "body                                      0\n",
       "alchemy_category_score                    0\n",
       "avgLinkWordLength                         0\n",
       "AvglinkWithOneCommonWord                  0\n",
       "AvglinkWithTwoCommonWord                  0\n",
       "AvglinkWithThreeCommonWord                0\n",
       "AvglinkWithFourCommonWord                 0\n",
       "redundancyMeasure                         0\n",
       "frameTagRatio                             0\n",
       "tagRatio                                  0\n",
       "imageTagRatio                             0\n",
       "isNews                                    0\n",
       "lengthyDomain                             0\n",
       "hyperlinkToAllWordsRatio                  0\n",
       "isFrontPageNews                           0\n",
       "alphanumCharCount                         0\n",
       "linksCount                                0\n",
       "wordCount                                 0\n",
       "parametrizedLinkRatio                     0\n",
       "spellingErrorsRatio                       0\n",
       "label                                  1479\n",
       "id                                        0\n",
       "alchemy_category_arts_entertainment       0\n",
       "alchemy_category_business                 0\n",
       "alchemy_category_computer_internet        0\n",
       "alchemy_category_culture_politics         0\n",
       "alchemy_category_gaming                   0\n",
       "alchemy_category_health                   0\n",
       "alchemy_category_law_crime                0\n",
       "alchemy_category_recreation               0\n",
       "alchemy_category_religion                 0\n",
       "alchemy_category_science_technology       0\n",
       "alchemy_category_sports                   0\n",
       "alchemy_category_unknown                  0\n",
       "alchemy_category_weather                  0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df.isna().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "37921cce",
   "metadata": {},
   "source": [
    "## 1.8 NLP "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6c967fcb",
   "metadata": {},
   "source": [
    "### 1.8.1. Removing Punctuations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "ccfaedbd",
   "metadata": {},
   "outputs": [],
   "source": [
    "def remove_punct(text):\n",
    "    text_nopunct = \"\".join([c for c in text if c not in string.punctuation])\n",
    "    return text_nopunct\n",
    "\n",
    "body = main_df[\"body\"].to_numpy()\n",
    "body_chan = []\n",
    "for i in range(len(body)):\n",
    "    body[i] = body[i].strip()\n",
    "    body_chan.append( remove_punct(body[i]))\n",
    "    \n",
    "main_df[\"body\"] = body_chan\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "26b1baf2",
   "metadata": {},
   "outputs": [],
   "source": [
    "body = main_df[\"title\"].to_numpy()\n",
    "body_chan = []\n",
    "for i in range(len(body)):\n",
    "    body[i] = body[i].strip()\n",
    "    body_chan.append( remove_punct(body[i]))\n",
    "    \n",
    "main_df[\"title\"] = body_chan"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "16e5fee9",
   "metadata": {},
   "outputs": [],
   "source": [
    "body = main_df[\"url\"].to_numpy()\n",
    "body_chan = []\n",
    "for i in range(len(body)):\n",
    "    body[i] = body[i].strip()\n",
    "    body_chan.append( remove_punct(body[i]))\n",
    "    \n",
    "main_df[\"url\"] = body_chan"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "id": "c2e8ac2e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>url</th>\n",
       "      <th>title</th>\n",
       "      <th>body</th>\n",
       "      <th>alchemy_category_score</th>\n",
       "      <th>avgLinkWordLength</th>\n",
       "      <th>AvglinkWithOneCommonWord</th>\n",
       "      <th>AvglinkWithTwoCommonWord</th>\n",
       "      <th>AvglinkWithThreeCommonWord</th>\n",
       "      <th>AvglinkWithFourCommonWord</th>\n",
       "      <th>redundancyMeasure</th>\n",
       "      <th>...</th>\n",
       "      <th>alchemy_category_culture_politics</th>\n",
       "      <th>alchemy_category_gaming</th>\n",
       "      <th>alchemy_category_health</th>\n",
       "      <th>alchemy_category_law_crime</th>\n",
       "      <th>alchemy_category_recreation</th>\n",
       "      <th>alchemy_category_religion</th>\n",
       "      <th>alchemy_category_science_technology</th>\n",
       "      <th>alchemy_category_sports</th>\n",
       "      <th>alchemy_category_unknown</th>\n",
       "      <th>alchemy_category_weather</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>polyvore cgi home id 1389651</td>\n",
       "      <td>Join Polyvore and follow the latest styles</td>\n",
       "      <td>polyvore is the best place to discover or star...</td>\n",
       "      <td>1.869397</td>\n",
       "      <td>1.070441</td>\n",
       "      <td>2.850855</td>\n",
       "      <td>0.007905</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.589894</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>youtube watch v ippMPPu6gh4</td>\n",
       "      <td>Speed Air ManDavid Belle</td>\n",
       "      <td>Speed Air ManDavid Belle david belle speed air...</td>\n",
       "      <td>1.869397</td>\n",
       "      <td>0.814292</td>\n",
       "      <td>3.129977</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.591068</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>musingsofahousewife 2011 03 tried and true chi...</td>\n",
       "      <td>Chicken Gruyere Recipe</td>\n",
       "      <td>Chicken Gruyere one of our favorite special di...</td>\n",
       "      <td>1.472093</td>\n",
       "      <td>1.106580</td>\n",
       "      <td>5.105662</td>\n",
       "      <td>0.250017</td>\n",
       "      <td>0.209826</td>\n",
       "      <td>0.184265</td>\n",
       "      <td>0.367008</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>thelittleteochew 2011 07 ikan bilis with tempe...</td>\n",
       "      <td>Ikan Bilis With Tempeh Groundnuts</td>\n",
       "      <td>Oh me oh my This was really snackalicious swee...</td>\n",
       "      <td>1.608077</td>\n",
       "      <td>0.980299</td>\n",
       "      <td>4.135385</td>\n",
       "      <td>0.181739</td>\n",
       "      <td>0.093401</td>\n",
       "      <td>0.064320</td>\n",
       "      <td>0.387063</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>recipes wuzzle index php 72</td>\n",
       "      <td>Malaysian recipes</td>\n",
       "      <td>Barbecued Chicken Chow Siew from The Exotic Ki...</td>\n",
       "      <td>1.707464</td>\n",
       "      <td>0.457966</td>\n",
       "      <td>2.818948</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.256666</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 37 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 url  \\\n",
       "0                       polyvore cgi home id 1389651   \n",
       "1                        youtube watch v ippMPPu6gh4   \n",
       "2  musingsofahousewife 2011 03 tried and true chi...   \n",
       "3  thelittleteochew 2011 07 ikan bilis with tempe...   \n",
       "4                        recipes wuzzle index php 72   \n",
       "\n",
       "                                        title  \\\n",
       "0  Join Polyvore and follow the latest styles   \n",
       "1                    Speed Air ManDavid Belle   \n",
       "2                      Chicken Gruyere Recipe   \n",
       "3           Ikan Bilis With Tempeh Groundnuts   \n",
       "4                           Malaysian recipes   \n",
       "\n",
       "                                                body  alchemy_category_score  \\\n",
       "0  polyvore is the best place to discover or star...                1.869397   \n",
       "1  Speed Air ManDavid Belle david belle speed air...                1.869397   \n",
       "2  Chicken Gruyere one of our favorite special di...                1.472093   \n",
       "3  Oh me oh my This was really snackalicious swee...                1.608077   \n",
       "4  Barbecued Chicken Chow Siew from The Exotic Ki...                1.707464   \n",
       "\n",
       "   avgLinkWordLength  AvglinkWithOneCommonWord  AvglinkWithTwoCommonWord  \\\n",
       "0           1.070441                  2.850855                  0.007905   \n",
       "1           0.814292                  3.129977                  0.000000   \n",
       "2           1.106580                  5.105662                  0.250017   \n",
       "3           0.980299                  4.135385                  0.181739   \n",
       "4           0.457966                  2.818948                  0.000000   \n",
       "\n",
       "   AvglinkWithThreeCommonWord  AvglinkWithFourCommonWord  redundancyMeasure  \\\n",
       "0                    0.000000                   0.000000           0.589894   \n",
       "1                    0.000000                   0.000000           0.591068   \n",
       "2                    0.209826                   0.184265           0.367008   \n",
       "3                    0.093401                   0.064320           0.387063   \n",
       "4                    0.000000                   0.000000           0.256666   \n",
       "\n",
       "   ...  alchemy_category_culture_politics  alchemy_category_gaming  \\\n",
       "0  ...                                  0                        0   \n",
       "1  ...                                  0                        0   \n",
       "2  ...                                  0                        0   \n",
       "3  ...                                  0                        0   \n",
       "4  ...                                  0                        0   \n",
       "\n",
       "   alchemy_category_health alchemy_category_law_crime  \\\n",
       "0                        0                          0   \n",
       "1                        0                          0   \n",
       "2                        0                          0   \n",
       "3                        0                          0   \n",
       "4                        0                          0   \n",
       "\n",
       "   alchemy_category_recreation  alchemy_category_religion  \\\n",
       "0                            0                          0   \n",
       "1                            0                          0   \n",
       "2                            0                          0   \n",
       "3                            1                          0   \n",
       "4                            0                          0   \n",
       "\n",
       "   alchemy_category_science_technology  alchemy_category_sports  \\\n",
       "0                                    0                        0   \n",
       "1                                    0                        0   \n",
       "2                                    1                        0   \n",
       "3                                    0                        0   \n",
       "4                                    0                        0   \n",
       "\n",
       "   alchemy_category_unknown  alchemy_category_weather  \n",
       "0                         1                         0  \n",
       "1                         1                         0  \n",
       "2                         0                         0  \n",
       "3                         0                         0  \n",
       "4                         0                         0  \n",
       "\n",
       "[5 rows x 37 columns]"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "994d9cde",
   "metadata": {},
   "source": [
    "### 1.8.2. Tokenization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "id": "f226cff8",
   "metadata": {},
   "outputs": [],
   "source": [
    "def tokenize(text):\n",
    "    tokens = re.split('\\W+',text)\n",
    "    return tokens\n",
    "\n",
    "token_change = []\n",
    "tok = main_df[\"body\"].to_numpy()\n",
    "for i in range(len(tok)):\n",
    "    token_change.append(tokenize(tok[i]))\n",
    "    \n",
    "main_df[\"body\"] = token_change  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "id": "74762887",
   "metadata": {},
   "outputs": [],
   "source": [
    "token_change = []\n",
    "tok = main_df[\"title\"].to_numpy()\n",
    "for i in range(len(tok)):\n",
    "    token_change.append(tokenize(tok[i]))\n",
    "    \n",
    "main_df[\"title\"] = token_change  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "id": "007b58f4",
   "metadata": {},
   "outputs": [],
   "source": [
    "token_change = []\n",
    "tok = main_df[\"url\"].to_numpy()\n",
    "for i in range(len(tok)):\n",
    "    token_change.append(tokenize(tok[i]))\n",
    "    \n",
    "main_df[\"url\"] = token_change  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "id": "112f62a5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>url</th>\n",
       "      <th>title</th>\n",
       "      <th>body</th>\n",
       "      <th>alchemy_category_score</th>\n",
       "      <th>avgLinkWordLength</th>\n",
       "      <th>AvglinkWithOneCommonWord</th>\n",
       "      <th>AvglinkWithTwoCommonWord</th>\n",
       "      <th>AvglinkWithThreeCommonWord</th>\n",
       "      <th>AvglinkWithFourCommonWord</th>\n",
       "      <th>redundancyMeasure</th>\n",
       "      <th>...</th>\n",
       "      <th>alchemy_category_culture_politics</th>\n",
       "      <th>alchemy_category_gaming</th>\n",
       "      <th>alchemy_category_health</th>\n",
       "      <th>alchemy_category_law_crime</th>\n",
       "      <th>alchemy_category_recreation</th>\n",
       "      <th>alchemy_category_religion</th>\n",
       "      <th>alchemy_category_science_technology</th>\n",
       "      <th>alchemy_category_sports</th>\n",
       "      <th>alchemy_category_unknown</th>\n",
       "      <th>alchemy_category_weather</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[polyvore, cgi, home, id, 1389651]</td>\n",
       "      <td>[Join, Polyvore, and, follow, the, latest, sty...</td>\n",
       "      <td>[polyvore, is, the, best, place, to, discover,...</td>\n",
       "      <td>1.869397</td>\n",
       "      <td>1.070441</td>\n",
       "      <td>2.850855</td>\n",
       "      <td>0.007905</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.589894</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[youtube, watch, v, ippMPPu6gh4]</td>\n",
       "      <td>[Speed, Air, ManDavid, Belle]</td>\n",
       "      <td>[Speed, Air, ManDavid, Belle, david, belle, sp...</td>\n",
       "      <td>1.869397</td>\n",
       "      <td>0.814292</td>\n",
       "      <td>3.129977</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.591068</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[musingsofahousewife, 2011, 03, tried, and, tr...</td>\n",
       "      <td>[Chicken, Gruyere, Recipe]</td>\n",
       "      <td>[Chicken, Gruyere, one, of, our, favorite, spe...</td>\n",
       "      <td>1.472093</td>\n",
       "      <td>1.106580</td>\n",
       "      <td>5.105662</td>\n",
       "      <td>0.250017</td>\n",
       "      <td>0.209826</td>\n",
       "      <td>0.184265</td>\n",
       "      <td>0.367008</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[thelittleteochew, 2011, 07, ikan, bilis, with...</td>\n",
       "      <td>[Ikan, Bilis, With, Tempeh, Groundnuts]</td>\n",
       "      <td>[Oh, me, oh, my, This, was, really, snackalici...</td>\n",
       "      <td>1.608077</td>\n",
       "      <td>0.980299</td>\n",
       "      <td>4.135385</td>\n",
       "      <td>0.181739</td>\n",
       "      <td>0.093401</td>\n",
       "      <td>0.064320</td>\n",
       "      <td>0.387063</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[recipes, wuzzle, index, php, 72]</td>\n",
       "      <td>[Malaysian, recipes]</td>\n",
       "      <td>[Barbecued, Chicken, Chow, Siew, from, The, Ex...</td>\n",
       "      <td>1.707464</td>\n",
       "      <td>0.457966</td>\n",
       "      <td>2.818948</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.256666</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 37 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 url  \\\n",
       "0                 [polyvore, cgi, home, id, 1389651]   \n",
       "1                   [youtube, watch, v, ippMPPu6gh4]   \n",
       "2  [musingsofahousewife, 2011, 03, tried, and, tr...   \n",
       "3  [thelittleteochew, 2011, 07, ikan, bilis, with...   \n",
       "4                  [recipes, wuzzle, index, php, 72]   \n",
       "\n",
       "                                               title  \\\n",
       "0  [Join, Polyvore, and, follow, the, latest, sty...   \n",
       "1                      [Speed, Air, ManDavid, Belle]   \n",
       "2                         [Chicken, Gruyere, Recipe]   \n",
       "3            [Ikan, Bilis, With, Tempeh, Groundnuts]   \n",
       "4                               [Malaysian, recipes]   \n",
       "\n",
       "                                                body  alchemy_category_score  \\\n",
       "0  [polyvore, is, the, best, place, to, discover,...                1.869397   \n",
       "1  [Speed, Air, ManDavid, Belle, david, belle, sp...                1.869397   \n",
       "2  [Chicken, Gruyere, one, of, our, favorite, spe...                1.472093   \n",
       "3  [Oh, me, oh, my, This, was, really, snackalici...                1.608077   \n",
       "4  [Barbecued, Chicken, Chow, Siew, from, The, Ex...                1.707464   \n",
       "\n",
       "   avgLinkWordLength  AvglinkWithOneCommonWord  AvglinkWithTwoCommonWord  \\\n",
       "0           1.070441                  2.850855                  0.007905   \n",
       "1           0.814292                  3.129977                  0.000000   \n",
       "2           1.106580                  5.105662                  0.250017   \n",
       "3           0.980299                  4.135385                  0.181739   \n",
       "4           0.457966                  2.818948                  0.000000   \n",
       "\n",
       "   AvglinkWithThreeCommonWord  AvglinkWithFourCommonWord  redundancyMeasure  \\\n",
       "0                    0.000000                   0.000000           0.589894   \n",
       "1                    0.000000                   0.000000           0.591068   \n",
       "2                    0.209826                   0.184265           0.367008   \n",
       "3                    0.093401                   0.064320           0.387063   \n",
       "4                    0.000000                   0.000000           0.256666   \n",
       "\n",
       "   ...  alchemy_category_culture_politics  alchemy_category_gaming  \\\n",
       "0  ...                                  0                        0   \n",
       "1  ...                                  0                        0   \n",
       "2  ...                                  0                        0   \n",
       "3  ...                                  0                        0   \n",
       "4  ...                                  0                        0   \n",
       "\n",
       "   alchemy_category_health alchemy_category_law_crime  \\\n",
       "0                        0                          0   \n",
       "1                        0                          0   \n",
       "2                        0                          0   \n",
       "3                        0                          0   \n",
       "4                        0                          0   \n",
       "\n",
       "   alchemy_category_recreation  alchemy_category_religion  \\\n",
       "0                            0                          0   \n",
       "1                            0                          0   \n",
       "2                            0                          0   \n",
       "3                            1                          0   \n",
       "4                            0                          0   \n",
       "\n",
       "   alchemy_category_science_technology  alchemy_category_sports  \\\n",
       "0                                    0                        0   \n",
       "1                                    0                        0   \n",
       "2                                    1                        0   \n",
       "3                                    0                        0   \n",
       "4                                    0                        0   \n",
       "\n",
       "   alchemy_category_unknown  alchemy_category_weather  \n",
       "0                         1                         0  \n",
       "1                         1                         0  \n",
       "2                         0                         0  \n",
       "3                         0                         0  \n",
       "4                         0                         0  \n",
       "\n",
       "[5 rows x 37 columns]"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c15bf69a",
   "metadata": {},
   "source": [
    "### 1.8.3. Remove Stop words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "id": "f64f808b",
   "metadata": {},
   "outputs": [],
   "source": [
    "stopwords = nltk.corpus.stopwords.words('english')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "id": "e8ddc2d6",
   "metadata": {},
   "outputs": [],
   "source": [
    "def remove_stopwords(tokenized_list):\n",
    "    text = [word for word in tokenized_list if word not in stopwords]\n",
    "    return text\n",
    "\n",
    "stop_list = main_df[\"body\"].to_numpy()\n",
    "stop_list_change = []\n",
    "for i in range(len(stop_list)):\n",
    "    stop_list_change.append(remove_stopwords(stop_list[i]))\n",
    "    \n",
    "main_df[\"body\"] = stop_list_change"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "id": "7db7b96e",
   "metadata": {},
   "outputs": [],
   "source": [
    "stop_list = main_df[\"title\"].to_numpy()\n",
    "stop_list_change = []\n",
    "for i in range(len(stop_list)):\n",
    "    stop_list_change.append(remove_stopwords(stop_list[i]))\n",
    "    \n",
    "main_df[\"title\"] = stop_list_change"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "id": "dcdd9ac9",
   "metadata": {},
   "outputs": [],
   "source": [
    "stop_list = main_df[\"url\"].to_numpy()\n",
    "stop_list_change = []\n",
    "for i in range(len(stop_list)):\n",
    "    stop_list_change.append(remove_stopwords(stop_list[i]))\n",
    "    \n",
    "main_df[\"url\"] = stop_list_change"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "id": "260ea178",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>url</th>\n",
       "      <th>title</th>\n",
       "      <th>body</th>\n",
       "      <th>alchemy_category_score</th>\n",
       "      <th>avgLinkWordLength</th>\n",
       "      <th>AvglinkWithOneCommonWord</th>\n",
       "      <th>AvglinkWithTwoCommonWord</th>\n",
       "      <th>AvglinkWithThreeCommonWord</th>\n",
       "      <th>AvglinkWithFourCommonWord</th>\n",
       "      <th>redundancyMeasure</th>\n",
       "      <th>...</th>\n",
       "      <th>alchemy_category_culture_politics</th>\n",
       "      <th>alchemy_category_gaming</th>\n",
       "      <th>alchemy_category_health</th>\n",
       "      <th>alchemy_category_law_crime</th>\n",
       "      <th>alchemy_category_recreation</th>\n",
       "      <th>alchemy_category_religion</th>\n",
       "      <th>alchemy_category_science_technology</th>\n",
       "      <th>alchemy_category_sports</th>\n",
       "      <th>alchemy_category_unknown</th>\n",
       "      <th>alchemy_category_weather</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[polyvore, cgi, home, id, 1389651]</td>\n",
       "      <td>[Join, Polyvore, follow, latest, styles]</td>\n",
       "      <td>[polyvore, best, place, discover, start, fashi...</td>\n",
       "      <td>1.869397</td>\n",
       "      <td>1.070441</td>\n",
       "      <td>2.850855</td>\n",
       "      <td>0.007905</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.589894</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[youtube, watch, v, ippMPPu6gh4]</td>\n",
       "      <td>[Speed, Air, ManDavid, Belle]</td>\n",
       "      <td>[Speed, Air, ManDavid, Belle, david, belle, sp...</td>\n",
       "      <td>1.869397</td>\n",
       "      <td>0.814292</td>\n",
       "      <td>3.129977</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.591068</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[musingsofahousewife, 2011, 03, tried, true, c...</td>\n",
       "      <td>[Chicken, Gruyere, Recipe]</td>\n",
       "      <td>[Chicken, Gruyere, one, favorite, special, din...</td>\n",
       "      <td>1.472093</td>\n",
       "      <td>1.106580</td>\n",
       "      <td>5.105662</td>\n",
       "      <td>0.250017</td>\n",
       "      <td>0.209826</td>\n",
       "      <td>0.184265</td>\n",
       "      <td>0.367008</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[thelittleteochew, 2011, 07, ikan, bilis, temp...</td>\n",
       "      <td>[Ikan, Bilis, With, Tempeh, Groundnuts]</td>\n",
       "      <td>[Oh, oh, This, really, snackalicious, sweet, s...</td>\n",
       "      <td>1.608077</td>\n",
       "      <td>0.980299</td>\n",
       "      <td>4.135385</td>\n",
       "      <td>0.181739</td>\n",
       "      <td>0.093401</td>\n",
       "      <td>0.064320</td>\n",
       "      <td>0.387063</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[recipes, wuzzle, index, php, 72]</td>\n",
       "      <td>[Malaysian, recipes]</td>\n",
       "      <td>[Barbecued, Chicken, Chow, Siew, The, Exotic, ...</td>\n",
       "      <td>1.707464</td>\n",
       "      <td>0.457966</td>\n",
       "      <td>2.818948</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.256666</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 37 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 url  \\\n",
       "0                 [polyvore, cgi, home, id, 1389651]   \n",
       "1                   [youtube, watch, v, ippMPPu6gh4]   \n",
       "2  [musingsofahousewife, 2011, 03, tried, true, c...   \n",
       "3  [thelittleteochew, 2011, 07, ikan, bilis, temp...   \n",
       "4                  [recipes, wuzzle, index, php, 72]   \n",
       "\n",
       "                                      title  \\\n",
       "0  [Join, Polyvore, follow, latest, styles]   \n",
       "1             [Speed, Air, ManDavid, Belle]   \n",
       "2                [Chicken, Gruyere, Recipe]   \n",
       "3   [Ikan, Bilis, With, Tempeh, Groundnuts]   \n",
       "4                      [Malaysian, recipes]   \n",
       "\n",
       "                                                body  alchemy_category_score  \\\n",
       "0  [polyvore, best, place, discover, start, fashi...                1.869397   \n",
       "1  [Speed, Air, ManDavid, Belle, david, belle, sp...                1.869397   \n",
       "2  [Chicken, Gruyere, one, favorite, special, din...                1.472093   \n",
       "3  [Oh, oh, This, really, snackalicious, sweet, s...                1.608077   \n",
       "4  [Barbecued, Chicken, Chow, Siew, The, Exotic, ...                1.707464   \n",
       "\n",
       "   avgLinkWordLength  AvglinkWithOneCommonWord  AvglinkWithTwoCommonWord  \\\n",
       "0           1.070441                  2.850855                  0.007905   \n",
       "1           0.814292                  3.129977                  0.000000   \n",
       "2           1.106580                  5.105662                  0.250017   \n",
       "3           0.980299                  4.135385                  0.181739   \n",
       "4           0.457966                  2.818948                  0.000000   \n",
       "\n",
       "   AvglinkWithThreeCommonWord  AvglinkWithFourCommonWord  redundancyMeasure  \\\n",
       "0                    0.000000                   0.000000           0.589894   \n",
       "1                    0.000000                   0.000000           0.591068   \n",
       "2                    0.209826                   0.184265           0.367008   \n",
       "3                    0.093401                   0.064320           0.387063   \n",
       "4                    0.000000                   0.000000           0.256666   \n",
       "\n",
       "   ...  alchemy_category_culture_politics  alchemy_category_gaming  \\\n",
       "0  ...                                  0                        0   \n",
       "1  ...                                  0                        0   \n",
       "2  ...                                  0                        0   \n",
       "3  ...                                  0                        0   \n",
       "4  ...                                  0                        0   \n",
       "\n",
       "   alchemy_category_health alchemy_category_law_crime  \\\n",
       "0                        0                          0   \n",
       "1                        0                          0   \n",
       "2                        0                          0   \n",
       "3                        0                          0   \n",
       "4                        0                          0   \n",
       "\n",
       "   alchemy_category_recreation  alchemy_category_religion  \\\n",
       "0                            0                          0   \n",
       "1                            0                          0   \n",
       "2                            0                          0   \n",
       "3                            1                          0   \n",
       "4                            0                          0   \n",
       "\n",
       "   alchemy_category_science_technology  alchemy_category_sports  \\\n",
       "0                                    0                        0   \n",
       "1                                    0                        0   \n",
       "2                                    1                        0   \n",
       "3                                    0                        0   \n",
       "4                                    0                        0   \n",
       "\n",
       "   alchemy_category_unknown  alchemy_category_weather  \n",
       "0                         1                         0  \n",
       "1                         1                         0  \n",
       "2                         0                         0  \n",
       "3                         0                         0  \n",
       "4                         0                         0  \n",
       "\n",
       "[5 rows x 37 columns]"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f52f75d5",
   "metadata": {},
   "source": [
    "### 1.8.4. Lemmatizing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "id": "0a097752",
   "metadata": {},
   "outputs": [],
   "source": [
    "wnl = nltk.WordNetLemmatizer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "id": "0211f988",
   "metadata": {},
   "outputs": [],
   "source": [
    "def lemmatizing(tokenized_text):\n",
    "    text = [wnl.lemmatize(word) for word in tokenized_text]\n",
    "    return text"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "id": "d68bdc4a",
   "metadata": {},
   "outputs": [],
   "source": [
    "lemmatize_list = main_df[\"url\"].to_numpy()\n",
    "lemmatized_list = []\n",
    "for i in range(len(lemmatize_list)):\n",
    "    lemmatized_list.append(lemmatizing(lemmatize_list[i]))\n",
    "    \n",
    "main_df[\"url\"] = lemmatized_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "id": "a7b82ec2",
   "metadata": {},
   "outputs": [],
   "source": [
    "lemmatize_list = main_df[\"title\"].to_numpy()\n",
    "lemmatized_list = []\n",
    "for i in range(len(lemmatize_list)):\n",
    "    lemmatized_list.append(lemmatizing(lemmatize_list[i]))\n",
    "    \n",
    "main_df[\"title\"] = lemmatized_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "id": "32e5b2f6",
   "metadata": {},
   "outputs": [],
   "source": [
    "lemmatize_list = main_df[\"body\"].to_numpy()\n",
    "lemmatized_list = []\n",
    "for i in range(len(lemmatize_list)):\n",
    "    lemmatized_list.append(lemmatizing(lemmatize_list[i]))\n",
    "    \n",
    "main_df[\"body\"] = lemmatized_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "id": "18fde14e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# st = nltk.PorterStemmer() "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "id": "afa1d217",
   "metadata": {},
   "outputs": [],
   "source": [
    "# def stemming(tokenized_text):\n",
    "#     text = [st.stem(word) for word in tokenized_text]\n",
    "#     return text"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "id": "8e71f0f7",
   "metadata": {},
   "outputs": [],
   "source": [
    "# stem_list = main_df[\"url\"].to_numpy()\n",
    "# stemmed_list = []\n",
    "# for i in range(len(stem_list)):\n",
    "#     stemmed_list.append(stemming(stem_list[i]))\n",
    "    \n",
    "# main_df[\"url\"] = stemmed_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "id": "68698cfe",
   "metadata": {},
   "outputs": [],
   "source": [
    "# stem_list = main_df[\"title\"].to_numpy()\n",
    "# stemmed_list = []\n",
    "# for i in range(len(stem_list)):\n",
    "#     stemmed_list.append(stemming(stem_list[i]))\n",
    "    \n",
    "# main_df[\"title\"] = stemmed_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "id": "04e3ce06",
   "metadata": {},
   "outputs": [],
   "source": [
    "# stem_list = main_df[\"body\"].to_numpy()\n",
    "# stemmed_list = []\n",
    "# for i in range(len(stem_list)):\n",
    "#     stemmed_list.append(stemming(stem_list[i]))\n",
    "    \n",
    "# main_df[\"body\"] = stemmed_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "id": "e1d60c7b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>url</th>\n",
       "      <th>title</th>\n",
       "      <th>body</th>\n",
       "      <th>alchemy_category_score</th>\n",
       "      <th>avgLinkWordLength</th>\n",
       "      <th>AvglinkWithOneCommonWord</th>\n",
       "      <th>AvglinkWithTwoCommonWord</th>\n",
       "      <th>AvglinkWithThreeCommonWord</th>\n",
       "      <th>AvglinkWithFourCommonWord</th>\n",
       "      <th>redundancyMeasure</th>\n",
       "      <th>...</th>\n",
       "      <th>alchemy_category_culture_politics</th>\n",
       "      <th>alchemy_category_gaming</th>\n",
       "      <th>alchemy_category_health</th>\n",
       "      <th>alchemy_category_law_crime</th>\n",
       "      <th>alchemy_category_recreation</th>\n",
       "      <th>alchemy_category_religion</th>\n",
       "      <th>alchemy_category_science_technology</th>\n",
       "      <th>alchemy_category_sports</th>\n",
       "      <th>alchemy_category_unknown</th>\n",
       "      <th>alchemy_category_weather</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[polyvore, cgi, home, id, 1389651]</td>\n",
       "      <td>[Join, Polyvore, follow, latest, style]</td>\n",
       "      <td>[polyvore, best, place, discover, start, fashi...</td>\n",
       "      <td>1.869397</td>\n",
       "      <td>1.070441</td>\n",
       "      <td>2.850855</td>\n",
       "      <td>0.007905</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.589894</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[youtube, watch, v, ippMPPu6gh4]</td>\n",
       "      <td>[Speed, Air, ManDavid, Belle]</td>\n",
       "      <td>[Speed, Air, ManDavid, Belle, david, belle, sp...</td>\n",
       "      <td>1.869397</td>\n",
       "      <td>0.814292</td>\n",
       "      <td>3.129977</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.591068</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[musingsofahousewife, 2011, 03, tried, true, c...</td>\n",
       "      <td>[Chicken, Gruyere, Recipe]</td>\n",
       "      <td>[Chicken, Gruyere, one, favorite, special, din...</td>\n",
       "      <td>1.472093</td>\n",
       "      <td>1.106580</td>\n",
       "      <td>5.105662</td>\n",
       "      <td>0.250017</td>\n",
       "      <td>0.209826</td>\n",
       "      <td>0.184265</td>\n",
       "      <td>0.367008</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[thelittleteochew, 2011, 07, ikan, bilis, temp...</td>\n",
       "      <td>[Ikan, Bilis, With, Tempeh, Groundnuts]</td>\n",
       "      <td>[Oh, oh, This, really, snackalicious, sweet, s...</td>\n",
       "      <td>1.608077</td>\n",
       "      <td>0.980299</td>\n",
       "      <td>4.135385</td>\n",
       "      <td>0.181739</td>\n",
       "      <td>0.093401</td>\n",
       "      <td>0.064320</td>\n",
       "      <td>0.387063</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[recipe, wuzzle, index, php, 72]</td>\n",
       "      <td>[Malaysian, recipe]</td>\n",
       "      <td>[Barbecued, Chicken, Chow, Siew, The, Exotic, ...</td>\n",
       "      <td>1.707464</td>\n",
       "      <td>0.457966</td>\n",
       "      <td>2.818948</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.256666</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 37 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 url  \\\n",
       "0                 [polyvore, cgi, home, id, 1389651]   \n",
       "1                   [youtube, watch, v, ippMPPu6gh4]   \n",
       "2  [musingsofahousewife, 2011, 03, tried, true, c...   \n",
       "3  [thelittleteochew, 2011, 07, ikan, bilis, temp...   \n",
       "4                   [recipe, wuzzle, index, php, 72]   \n",
       "\n",
       "                                     title  \\\n",
       "0  [Join, Polyvore, follow, latest, style]   \n",
       "1            [Speed, Air, ManDavid, Belle]   \n",
       "2               [Chicken, Gruyere, Recipe]   \n",
       "3  [Ikan, Bilis, With, Tempeh, Groundnuts]   \n",
       "4                      [Malaysian, recipe]   \n",
       "\n",
       "                                                body  alchemy_category_score  \\\n",
       "0  [polyvore, best, place, discover, start, fashi...                1.869397   \n",
       "1  [Speed, Air, ManDavid, Belle, david, belle, sp...                1.869397   \n",
       "2  [Chicken, Gruyere, one, favorite, special, din...                1.472093   \n",
       "3  [Oh, oh, This, really, snackalicious, sweet, s...                1.608077   \n",
       "4  [Barbecued, Chicken, Chow, Siew, The, Exotic, ...                1.707464   \n",
       "\n",
       "   avgLinkWordLength  AvglinkWithOneCommonWord  AvglinkWithTwoCommonWord  \\\n",
       "0           1.070441                  2.850855                  0.007905   \n",
       "1           0.814292                  3.129977                  0.000000   \n",
       "2           1.106580                  5.105662                  0.250017   \n",
       "3           0.980299                  4.135385                  0.181739   \n",
       "4           0.457966                  2.818948                  0.000000   \n",
       "\n",
       "   AvglinkWithThreeCommonWord  AvglinkWithFourCommonWord  redundancyMeasure  \\\n",
       "0                    0.000000                   0.000000           0.589894   \n",
       "1                    0.000000                   0.000000           0.591068   \n",
       "2                    0.209826                   0.184265           0.367008   \n",
       "3                    0.093401                   0.064320           0.387063   \n",
       "4                    0.000000                   0.000000           0.256666   \n",
       "\n",
       "   ...  alchemy_category_culture_politics  alchemy_category_gaming  \\\n",
       "0  ...                                  0                        0   \n",
       "1  ...                                  0                        0   \n",
       "2  ...                                  0                        0   \n",
       "3  ...                                  0                        0   \n",
       "4  ...                                  0                        0   \n",
       "\n",
       "   alchemy_category_health alchemy_category_law_crime  \\\n",
       "0                        0                          0   \n",
       "1                        0                          0   \n",
       "2                        0                          0   \n",
       "3                        0                          0   \n",
       "4                        0                          0   \n",
       "\n",
       "   alchemy_category_recreation  alchemy_category_religion  \\\n",
       "0                            0                          0   \n",
       "1                            0                          0   \n",
       "2                            0                          0   \n",
       "3                            1                          0   \n",
       "4                            0                          0   \n",
       "\n",
       "   alchemy_category_science_technology  alchemy_category_sports  \\\n",
       "0                                    0                        0   \n",
       "1                                    0                        0   \n",
       "2                                    1                        0   \n",
       "3                                    0                        0   \n",
       "4                                    0                        0   \n",
       "\n",
       "   alchemy_category_unknown  alchemy_category_weather  \n",
       "0                         1                         0  \n",
       "1                         1                         0  \n",
       "2                         0                         0  \n",
       "3                         0                         0  \n",
       "4                         0                         0  \n",
       "\n",
       "[5 rows x 37 columns]"
      ]
     },
     "execution_count": 101,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d76bff41",
   "metadata": {},
   "source": [
    "### 1.8.5. Vectorizing Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "id": "82546026",
   "metadata": {},
   "outputs": [],
   "source": [
    "body_data = main_df[\"body\"].to_numpy()\n",
    "body_data_combined = [' '.join(i) for i in body_data]\n",
    "main_df[\"body\"] = body_data_combined\n",
    "\n",
    "title_data = main_df[\"title\"].to_numpy()\n",
    "title_data_combined = [' '.join(i) for i in title_data]\n",
    "main_df[\"title\"] = title_data_combined\n",
    "\n",
    "url_data = main_df[\"url\"].to_numpy()\n",
    "url_data_combined = [' '.join(i) for i in url_data]\n",
    "main_df[\"url\"] = url_data_combined"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "id": "662b151a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>url</th>\n",
       "      <th>title</th>\n",
       "      <th>body</th>\n",
       "      <th>alchemy_category_score</th>\n",
       "      <th>avgLinkWordLength</th>\n",
       "      <th>AvglinkWithOneCommonWord</th>\n",
       "      <th>AvglinkWithTwoCommonWord</th>\n",
       "      <th>AvglinkWithThreeCommonWord</th>\n",
       "      <th>AvglinkWithFourCommonWord</th>\n",
       "      <th>redundancyMeasure</th>\n",
       "      <th>...</th>\n",
       "      <th>alchemy_category_culture_politics</th>\n",
       "      <th>alchemy_category_gaming</th>\n",
       "      <th>alchemy_category_health</th>\n",
       "      <th>alchemy_category_law_crime</th>\n",
       "      <th>alchemy_category_recreation</th>\n",
       "      <th>alchemy_category_religion</th>\n",
       "      <th>alchemy_category_science_technology</th>\n",
       "      <th>alchemy_category_sports</th>\n",
       "      <th>alchemy_category_unknown</th>\n",
       "      <th>alchemy_category_weather</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>polyvore cgi home id 1389651</td>\n",
       "      <td>Join Polyvore follow latest style</td>\n",
       "      <td>polyvore best place discover start fashion tre...</td>\n",
       "      <td>1.869397</td>\n",
       "      <td>1.070441</td>\n",
       "      <td>2.850855</td>\n",
       "      <td>0.007905</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.589894</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>youtube watch v ippMPPu6gh4</td>\n",
       "      <td>Speed Air ManDavid Belle</td>\n",
       "      <td>Speed Air ManDavid Belle david belle speed air...</td>\n",
       "      <td>1.869397</td>\n",
       "      <td>0.814292</td>\n",
       "      <td>3.129977</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.591068</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>musingsofahousewife 2011 03 tried true chicken...</td>\n",
       "      <td>Chicken Gruyere Recipe</td>\n",
       "      <td>Chicken Gruyere one favorite special dinner I ...</td>\n",
       "      <td>1.472093</td>\n",
       "      <td>1.106580</td>\n",
       "      <td>5.105662</td>\n",
       "      <td>0.250017</td>\n",
       "      <td>0.209826</td>\n",
       "      <td>0.184265</td>\n",
       "      <td>0.367008</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>thelittleteochew 2011 07 ikan bilis tempeh gro...</td>\n",
       "      <td>Ikan Bilis With Tempeh Groundnuts</td>\n",
       "      <td>Oh oh This really snackalicious sweet salty sp...</td>\n",
       "      <td>1.608077</td>\n",
       "      <td>0.980299</td>\n",
       "      <td>4.135385</td>\n",
       "      <td>0.181739</td>\n",
       "      <td>0.093401</td>\n",
       "      <td>0.064320</td>\n",
       "      <td>0.387063</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>recipe wuzzle index php 72</td>\n",
       "      <td>Malaysian recipe</td>\n",
       "      <td>Barbecued Chicken Chow Siew The Exotic Kitchen...</td>\n",
       "      <td>1.707464</td>\n",
       "      <td>0.457966</td>\n",
       "      <td>2.818948</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.256666</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 37 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 url  \\\n",
       "0                       polyvore cgi home id 1389651   \n",
       "1                        youtube watch v ippMPPu6gh4   \n",
       "2  musingsofahousewife 2011 03 tried true chicken...   \n",
       "3  thelittleteochew 2011 07 ikan bilis tempeh gro...   \n",
       "4                         recipe wuzzle index php 72   \n",
       "\n",
       "                               title  \\\n",
       "0  Join Polyvore follow latest style   \n",
       "1           Speed Air ManDavid Belle   \n",
       "2             Chicken Gruyere Recipe   \n",
       "3  Ikan Bilis With Tempeh Groundnuts   \n",
       "4                   Malaysian recipe   \n",
       "\n",
       "                                                body  alchemy_category_score  \\\n",
       "0  polyvore best place discover start fashion tre...                1.869397   \n",
       "1  Speed Air ManDavid Belle david belle speed air...                1.869397   \n",
       "2  Chicken Gruyere one favorite special dinner I ...                1.472093   \n",
       "3  Oh oh This really snackalicious sweet salty sp...                1.608077   \n",
       "4  Barbecued Chicken Chow Siew The Exotic Kitchen...                1.707464   \n",
       "\n",
       "   avgLinkWordLength  AvglinkWithOneCommonWord  AvglinkWithTwoCommonWord  \\\n",
       "0           1.070441                  2.850855                  0.007905   \n",
       "1           0.814292                  3.129977                  0.000000   \n",
       "2           1.106580                  5.105662                  0.250017   \n",
       "3           0.980299                  4.135385                  0.181739   \n",
       "4           0.457966                  2.818948                  0.000000   \n",
       "\n",
       "   AvglinkWithThreeCommonWord  AvglinkWithFourCommonWord  redundancyMeasure  \\\n",
       "0                    0.000000                   0.000000           0.589894   \n",
       "1                    0.000000                   0.000000           0.591068   \n",
       "2                    0.209826                   0.184265           0.367008   \n",
       "3                    0.093401                   0.064320           0.387063   \n",
       "4                    0.000000                   0.000000           0.256666   \n",
       "\n",
       "   ...  alchemy_category_culture_politics  alchemy_category_gaming  \\\n",
       "0  ...                                  0                        0   \n",
       "1  ...                                  0                        0   \n",
       "2  ...                                  0                        0   \n",
       "3  ...                                  0                        0   \n",
       "4  ...                                  0                        0   \n",
       "\n",
       "   alchemy_category_health alchemy_category_law_crime  \\\n",
       "0                        0                          0   \n",
       "1                        0                          0   \n",
       "2                        0                          0   \n",
       "3                        0                          0   \n",
       "4                        0                          0   \n",
       "\n",
       "   alchemy_category_recreation  alchemy_category_religion  \\\n",
       "0                            0                          0   \n",
       "1                            0                          0   \n",
       "2                            0                          0   \n",
       "3                            1                          0   \n",
       "4                            0                          0   \n",
       "\n",
       "   alchemy_category_science_technology  alchemy_category_sports  \\\n",
       "0                                    0                        0   \n",
       "1                                    0                        0   \n",
       "2                                    1                        0   \n",
       "3                                    0                        0   \n",
       "4                                    0                        0   \n",
       "\n",
       "   alchemy_category_unknown  alchemy_category_weather  \n",
       "0                         1                         0  \n",
       "1                         1                         0  \n",
       "2                         0                         0  \n",
       "3                         0                         0  \n",
       "4                         0                         0  \n",
       "\n",
       "[5 rows x 37 columns]"
      ]
     },
     "execution_count": 103,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "id": "d501ff55",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>alchemy_category_score</th>\n",
       "      <th>avgLinkWordLength</th>\n",
       "      <th>AvglinkWithOneCommonWord</th>\n",
       "      <th>AvglinkWithTwoCommonWord</th>\n",
       "      <th>AvglinkWithThreeCommonWord</th>\n",
       "      <th>AvglinkWithFourCommonWord</th>\n",
       "      <th>redundancyMeasure</th>\n",
       "      <th>frameTagRatio</th>\n",
       "      <th>tagRatio</th>\n",
       "      <th>imageTagRatio</th>\n",
       "      <th>...</th>\n",
       "      <th>alchemy_category_culture_politics</th>\n",
       "      <th>alchemy_category_gaming</th>\n",
       "      <th>alchemy_category_health</th>\n",
       "      <th>alchemy_category_law_crime</th>\n",
       "      <th>alchemy_category_recreation</th>\n",
       "      <th>alchemy_category_religion</th>\n",
       "      <th>alchemy_category_science_technology</th>\n",
       "      <th>alchemy_category_sports</th>\n",
       "      <th>alchemy_category_unknown</th>\n",
       "      <th>alchemy_category_weather</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.00000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>1.869713</td>\n",
       "      <td>1.128277</td>\n",
       "      <td>4.430076</td>\n",
       "      <td>0.186979</td>\n",
       "      <td>0.084568</td>\n",
       "      <td>0.045948</td>\n",
       "      <td>0.420341</td>\n",
       "      <td>0.054424</td>\n",
       "      <td>0.232441</td>\n",
       "      <td>0.137541</td>\n",
       "      <td>...</td>\n",
       "      <td>0.046383</td>\n",
       "      <td>0.010277</td>\n",
       "      <td>0.068425</td>\n",
       "      <td>0.004192</td>\n",
       "      <td>0.166193</td>\n",
       "      <td>0.009736</td>\n",
       "      <td>0.03908</td>\n",
       "      <td>0.051386</td>\n",
       "      <td>0.317512</td>\n",
       "      <td>0.000541</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.306335</td>\n",
       "      <td>0.281550</td>\n",
       "      <td>0.877527</td>\n",
       "      <td>0.117388</td>\n",
       "      <td>0.081797</td>\n",
       "      <td>0.063697</td>\n",
       "      <td>0.086000</td>\n",
       "      <td>0.034233</td>\n",
       "      <td>0.047146</td>\n",
       "      <td>0.223536</td>\n",
       "      <td>...</td>\n",
       "      <td>0.210327</td>\n",
       "      <td>0.100861</td>\n",
       "      <td>0.252490</td>\n",
       "      <td>0.064614</td>\n",
       "      <td>0.372279</td>\n",
       "      <td>0.098198</td>\n",
       "      <td>0.19380</td>\n",
       "      <td>0.220799</td>\n",
       "      <td>0.465540</td>\n",
       "      <td>0.023253</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.238949</td>\n",
       "      <td>0.457966</td>\n",
       "      <td>2.718282</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.231693</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.111499</td>\n",
       "      <td>-0.290300</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>1.713876</td>\n",
       "      <td>0.956304</td>\n",
       "      <td>3.820456</td>\n",
       "      <td>0.100083</td>\n",
       "      <td>0.021979</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.366458</td>\n",
       "      <td>0.028502</td>\n",
       "      <td>0.201061</td>\n",
       "      <td>0.025900</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>1.869397</td>\n",
       "      <td>1.127600</td>\n",
       "      <td>4.399459</td>\n",
       "      <td>0.184364</td>\n",
       "      <td>0.066375</td>\n",
       "      <td>0.021979</td>\n",
       "      <td>0.394525</td>\n",
       "      <td>0.045775</td>\n",
       "      <td>0.230564</td>\n",
       "      <td>0.083051</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>2.030494</td>\n",
       "      <td>1.288530</td>\n",
       "      <td>5.035959</td>\n",
       "      <td>0.262364</td>\n",
       "      <td>0.125163</td>\n",
       "      <td>0.063036</td>\n",
       "      <td>0.456302</td>\n",
       "      <td>0.073459</td>\n",
       "      <td>0.260770</td>\n",
       "      <td>0.236700</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2.505420</td>\n",
       "      <td>1.786869</td>\n",
       "      <td>7.389056</td>\n",
       "      <td>0.693147</td>\n",
       "      <td>0.683295</td>\n",
       "      <td>0.683295</td>\n",
       "      <td>0.591068</td>\n",
       "      <td>0.140893</td>\n",
       "      <td>0.350332</td>\n",
       "      <td>0.552900</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.00000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 33 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       alchemy_category_score  avgLinkWordLength  AvglinkWithOneCommonWord  \\\n",
       "count             7395.000000        7395.000000               7395.000000   \n",
       "mean                 1.869713           1.128277                  4.430076   \n",
       "std                  0.306335           0.281550                  0.877527   \n",
       "min                  1.238949           0.457966                  2.718282   \n",
       "25%                  1.713876           0.956304                  3.820456   \n",
       "50%                  1.869397           1.127600                  4.399459   \n",
       "75%                  2.030494           1.288530                  5.035959   \n",
       "max                  2.505420           1.786869                  7.389056   \n",
       "\n",
       "       AvglinkWithTwoCommonWord  AvglinkWithThreeCommonWord  \\\n",
       "count               7395.000000                 7395.000000   \n",
       "mean                   0.186979                    0.084568   \n",
       "std                    0.117388                    0.081797   \n",
       "min                    0.000000                    0.000000   \n",
       "25%                    0.100083                    0.021979   \n",
       "50%                    0.184364                    0.066375   \n",
       "75%                    0.262364                    0.125163   \n",
       "max                    0.693147                    0.683295   \n",
       "\n",
       "       AvglinkWithFourCommonWord  redundancyMeasure  frameTagRatio  \\\n",
       "count                7395.000000        7395.000000    7395.000000   \n",
       "mean                    0.045948           0.420341       0.054424   \n",
       "std                     0.063697           0.086000       0.034233   \n",
       "min                     0.000000           0.231693       0.000000   \n",
       "25%                     0.000000           0.366458       0.028502   \n",
       "50%                     0.021979           0.394525       0.045775   \n",
       "75%                     0.063036           0.456302       0.073459   \n",
       "max                     0.683295           0.591068       0.140893   \n",
       "\n",
       "          tagRatio  imageTagRatio  ...  alchemy_category_culture_politics  \\\n",
       "count  7395.000000    7395.000000  ...                        7395.000000   \n",
       "mean      0.232441       0.137541  ...                           0.046383   \n",
       "std       0.047146       0.223536  ...                           0.210327   \n",
       "min       0.111499      -0.290300  ...                           0.000000   \n",
       "25%       0.201061       0.025900  ...                           0.000000   \n",
       "50%       0.230564       0.083051  ...                           0.000000   \n",
       "75%       0.260770       0.236700  ...                           0.000000   \n",
       "max       0.350332       0.552900  ...                           1.000000   \n",
       "\n",
       "       alchemy_category_gaming  alchemy_category_health  \\\n",
       "count              7395.000000              7395.000000   \n",
       "mean                  0.010277                 0.068425   \n",
       "std                   0.100861                 0.252490   \n",
       "min                   0.000000                 0.000000   \n",
       "25%                   0.000000                 0.000000   \n",
       "50%                   0.000000                 0.000000   \n",
       "75%                   0.000000                 0.000000   \n",
       "max                   1.000000                 1.000000   \n",
       "\n",
       "       alchemy_category_law_crime  alchemy_category_recreation  \\\n",
       "count                 7395.000000                  7395.000000   \n",
       "mean                     0.004192                     0.166193   \n",
       "std                      0.064614                     0.372279   \n",
       "min                      0.000000                     0.000000   \n",
       "25%                      0.000000                     0.000000   \n",
       "50%                      0.000000                     0.000000   \n",
       "75%                      0.000000                     0.000000   \n",
       "max                      1.000000                     1.000000   \n",
       "\n",
       "       alchemy_category_religion  alchemy_category_science_technology  \\\n",
       "count                7395.000000                           7395.00000   \n",
       "mean                    0.009736                              0.03908   \n",
       "std                     0.098198                              0.19380   \n",
       "min                     0.000000                              0.00000   \n",
       "25%                     0.000000                              0.00000   \n",
       "50%                     0.000000                              0.00000   \n",
       "75%                     0.000000                              0.00000   \n",
       "max                     1.000000                              1.00000   \n",
       "\n",
       "       alchemy_category_sports  alchemy_category_unknown  \\\n",
       "count              7395.000000               7395.000000   \n",
       "mean                  0.051386                  0.317512   \n",
       "std                   0.220799                  0.465540   \n",
       "min                   0.000000                  0.000000   \n",
       "25%                   0.000000                  0.000000   \n",
       "50%                   0.000000                  0.000000   \n",
       "75%                   0.000000                  1.000000   \n",
       "max                   1.000000                  1.000000   \n",
       "\n",
       "       alchemy_category_weather  \n",
       "count               7395.000000  \n",
       "mean                   0.000541  \n",
       "std                    0.023253  \n",
       "min                    0.000000  \n",
       "25%                    0.000000  \n",
       "50%                    0.000000  \n",
       "75%                    0.000000  \n",
       "max                    1.000000  \n",
       "\n",
       "[8 rows x 33 columns]"
      ]
     },
     "execution_count": 104,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "id": "50e10cf2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0       3711\n",
       "1       7222\n",
       "2       3964\n",
       "3       3697\n",
       "4       4321\n",
       "        ... \n",
       "1474    3336\n",
       "1475    2867\n",
       "1476     738\n",
       "1477    4124\n",
       "1478    6056\n",
       "Name: id, Length: 7395, dtype: int64"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df[\"id\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "id": "4f09ae75",
   "metadata": {},
   "outputs": [],
   "source": [
    "main_df['label_out'] = main_df['label']\n",
    "main_df['id_out'] = main_df['id']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "id": "0080680c",
   "metadata": {},
   "outputs": [],
   "source": [
    "main_df[\"total_description\"] = main_df[\"title\"] + main_df[\"body\"] + main_df[\"url\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "id": "f29b4c31",
   "metadata": {},
   "outputs": [],
   "source": [
    "columns_to_be_deleted = [\"title\", \"body\",\"url\", \"label\",\"id\"] \n",
    "main_df.drop(axis=\"columns\", labels=columns_to_be_deleted, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5d35aae7",
   "metadata": {},
   "source": [
    "### Using Count Vectorizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "id": "9b1b08a8",
   "metadata": {},
   "outputs": [],
   "source": [
    "# cv = CountVectorizer(stop_words = {'english'},max_features = 300)\n",
    "# count_vector_body = cv.fit_transform(main_df[\"total_description\"])\n",
    "# print(count_vector_body.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "id": "06659560",
   "metadata": {},
   "outputs": [],
   "source": [
    "# names_list = cv.get_feature_names()\n",
    "# len(names_list)\n",
    "# count_vector_body = count_vector_body.todense() "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "019828d6",
   "metadata": {},
   "source": [
    "### Using Tfidf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "id": "389b41ad",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(7395, 3000)\n"
     ]
    }
   ],
   "source": [
    "tfidf_vect = TfidfVectorizer(stop_words = {'english'}, max_features = 3000)\n",
    "X_tfidf = tfidf_vect.fit_transform(main_df['total_description'])\n",
    "print(X_tfidf.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "id": "de4bdc33",
   "metadata": {},
   "outputs": [],
   "source": [
    "names_list = tfidf_vect.get_feature_names()\n",
    "len(names_list)\n",
    "tfidf = X_tfidf.todense() "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "id": "467f5548",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.DataFrame(tfidf, columns = names_list) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "id": "15fb792c",
   "metadata": {},
   "outputs": [],
   "source": [
    "main_df_list = main_df.to_numpy() \n",
    "column_names = main_df.columns\n",
    "main_df = pd.DataFrame(main_df_list, columns = column_names)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "id": "e27188cf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>alchemy_category_score</th>\n",
       "      <th>avgLinkWordLength</th>\n",
       "      <th>AvglinkWithOneCommonWord</th>\n",
       "      <th>AvglinkWithTwoCommonWord</th>\n",
       "      <th>AvglinkWithThreeCommonWord</th>\n",
       "      <th>AvglinkWithFourCommonWord</th>\n",
       "      <th>redundancyMeasure</th>\n",
       "      <th>frameTagRatio</th>\n",
       "      <th>tagRatio</th>\n",
       "      <th>imageTagRatio</th>\n",
       "      <th>...</th>\n",
       "      <th>alchemy_category_law_crime</th>\n",
       "      <th>alchemy_category_recreation</th>\n",
       "      <th>alchemy_category_religion</th>\n",
       "      <th>alchemy_category_science_technology</th>\n",
       "      <th>alchemy_category_sports</th>\n",
       "      <th>alchemy_category_unknown</th>\n",
       "      <th>alchemy_category_weather</th>\n",
       "      <th>label_out</th>\n",
       "      <th>id_out</th>\n",
       "      <th>total_description</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.869397</td>\n",
       "      <td>1.070441</td>\n",
       "      <td>2.850855</td>\n",
       "      <td>0.007905</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.589894</td>\n",
       "      <td>0.027778</td>\n",
       "      <td>0.203161</td>\n",
       "      <td>0.5529</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3711</td>\n",
       "      <td>Join Polyvore follow latest stylepolyvore best...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.869397</td>\n",
       "      <td>0.814292</td>\n",
       "      <td>3.129977</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.591068</td>\n",
       "      <td>0.015086</td>\n",
       "      <td>0.168189</td>\n",
       "      <td>0.5529</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>7222</td>\n",
       "      <td>Speed Air ManDavid BelleSpeed Air ManDavid Bel...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.472093</td>\n",
       "      <td>1.10658</td>\n",
       "      <td>5.105662</td>\n",
       "      <td>0.250017</td>\n",
       "      <td>0.209826</td>\n",
       "      <td>0.184265</td>\n",
       "      <td>0.367008</td>\n",
       "      <td>0.033935</td>\n",
       "      <td>0.192652</td>\n",
       "      <td>0.126582</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3964</td>\n",
       "      <td>Chicken Gruyere RecipeChicken Gruyere one favo...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.608077</td>\n",
       "      <td>0.980299</td>\n",
       "      <td>4.135385</td>\n",
       "      <td>0.181739</td>\n",
       "      <td>0.093401</td>\n",
       "      <td>0.06432</td>\n",
       "      <td>0.387063</td>\n",
       "      <td>0.03653</td>\n",
       "      <td>0.198</td>\n",
       "      <td>0.135593</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3697</td>\n",
       "      <td>Ikan Bilis With Tempeh GroundnutsOh oh This re...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.707464</td>\n",
       "      <td>0.457966</td>\n",
       "      <td>2.818948</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.256666</td>\n",
       "      <td>0.015152</td>\n",
       "      <td>0.241564</td>\n",
       "      <td>0.054217</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4321</td>\n",
       "      <td>Malaysian recipeBarbecued Chicken Chow Siew Th...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7390</th>\n",
       "      <td>1.976679</td>\n",
       "      <td>0.970779</td>\n",
       "      <td>3.925608</td>\n",
       "      <td>0.097638</td>\n",
       "      <td>0.033617</td>\n",
       "      <td>0.01695</td>\n",
       "      <td>0.591068</td>\n",
       "      <td>0.071719</td>\n",
       "      <td>0.237913</td>\n",
       "      <td>-0.2903</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3336</td>\n",
       "      <td>UnkonownUnkonownlanascooking 2011 07 08 roaste...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7391</th>\n",
       "      <td>1.869397</td>\n",
       "      <td>1.174511</td>\n",
       "      <td>5.348241</td>\n",
       "      <td>0.208901</td>\n",
       "      <td>0.087011</td>\n",
       "      <td>0.020001</td>\n",
       "      <td>0.345951</td>\n",
       "      <td>0.056471</td>\n",
       "      <td>0.237229</td>\n",
       "      <td>0.024793</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2867</td>\n",
       "      <td>Billy Chocolate Buttercream Martha Stewart Rec...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7392</th>\n",
       "      <td>2.50542</td>\n",
       "      <td>1.040202</td>\n",
       "      <td>4.538926</td>\n",
       "      <td>0.246225</td>\n",
       "      <td>0.054334</td>\n",
       "      <td>0.020101</td>\n",
       "      <td>0.436514</td>\n",
       "      <td>0.04129</td>\n",
       "      <td>0.21018</td>\n",
       "      <td>0.433526</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>738</td>\n",
       "      <td>Los Angeles Best Belly Dance Workout Swerve St...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7393</th>\n",
       "      <td>1.869397</td>\n",
       "      <td>1.009167</td>\n",
       "      <td>4.754497</td>\n",
       "      <td>0.204794</td>\n",
       "      <td>0.078643</td>\n",
       "      <td>0.018019</td>\n",
       "      <td>0.453249</td>\n",
       "      <td>0.070261</td>\n",
       "      <td>0.190801</td>\n",
       "      <td>0.5529</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4124</td>\n",
       "      <td>Buffalo Chicken Burgers recipe buffalo chicken...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7394</th>\n",
       "      <td>2.044258</td>\n",
       "      <td>1.105337</td>\n",
       "      <td>4.445692</td>\n",
       "      <td>0.196998</td>\n",
       "      <td>0.031749</td>\n",
       "      <td>0.027835</td>\n",
       "      <td>0.409636</td>\n",
       "      <td>0.012228</td>\n",
       "      <td>0.234657</td>\n",
       "      <td>0.078231</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>6056</td>\n",
       "      <td>canarygirl com Sirloin Peppercorn Cream SauceS...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>7395 rows × 35 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     alchemy_category_score avgLinkWordLength AvglinkWithOneCommonWord  \\\n",
       "0                  1.869397          1.070441                 2.850855   \n",
       "1                  1.869397          0.814292                 3.129977   \n",
       "2                  1.472093           1.10658                 5.105662   \n",
       "3                  1.608077          0.980299                 4.135385   \n",
       "4                  1.707464          0.457966                 2.818948   \n",
       "...                     ...               ...                      ...   \n",
       "7390               1.976679          0.970779                 3.925608   \n",
       "7391               1.869397          1.174511                 5.348241   \n",
       "7392                2.50542          1.040202                 4.538926   \n",
       "7393               1.869397          1.009167                 4.754497   \n",
       "7394               2.044258          1.105337                 4.445692   \n",
       "\n",
       "     AvglinkWithTwoCommonWord AvglinkWithThreeCommonWord  \\\n",
       "0                    0.007905                        0.0   \n",
       "1                         0.0                        0.0   \n",
       "2                    0.250017                   0.209826   \n",
       "3                    0.181739                   0.093401   \n",
       "4                         0.0                        0.0   \n",
       "...                       ...                        ...   \n",
       "7390                 0.097638                   0.033617   \n",
       "7391                 0.208901                   0.087011   \n",
       "7392                 0.246225                   0.054334   \n",
       "7393                 0.204794                   0.078643   \n",
       "7394                 0.196998                   0.031749   \n",
       "\n",
       "     AvglinkWithFourCommonWord redundancyMeasure frameTagRatio  tagRatio  \\\n",
       "0                          0.0          0.589894      0.027778  0.203161   \n",
       "1                          0.0          0.591068      0.015086  0.168189   \n",
       "2                     0.184265          0.367008      0.033935  0.192652   \n",
       "3                      0.06432          0.387063       0.03653     0.198   \n",
       "4                          0.0          0.256666      0.015152  0.241564   \n",
       "...                        ...               ...           ...       ...   \n",
       "7390                   0.01695          0.591068      0.071719  0.237913   \n",
       "7391                  0.020001          0.345951      0.056471  0.237229   \n",
       "7392                  0.020101          0.436514       0.04129   0.21018   \n",
       "7393                  0.018019          0.453249      0.070261  0.190801   \n",
       "7394                  0.027835          0.409636      0.012228  0.234657   \n",
       "\n",
       "     imageTagRatio  ... alchemy_category_law_crime  \\\n",
       "0           0.5529  ...                          0   \n",
       "1           0.5529  ...                          0   \n",
       "2         0.126582  ...                          0   \n",
       "3         0.135593  ...                          0   \n",
       "4         0.054217  ...                          0   \n",
       "...            ...  ...                        ...   \n",
       "7390       -0.2903  ...                          0   \n",
       "7391      0.024793  ...                          0   \n",
       "7392      0.433526  ...                          0   \n",
       "7393        0.5529  ...                          0   \n",
       "7394      0.078231  ...                          0   \n",
       "\n",
       "     alchemy_category_recreation alchemy_category_religion  \\\n",
       "0                              0                         0   \n",
       "1                              0                         0   \n",
       "2                              0                         0   \n",
       "3                              1                         0   \n",
       "4                              0                         0   \n",
       "...                          ...                       ...   \n",
       "7390                           0                         0   \n",
       "7391                           0                         0   \n",
       "7392                           0                         0   \n",
       "7393                           0                         0   \n",
       "7394                           0                         0   \n",
       "\n",
       "     alchemy_category_science_technology alchemy_category_sports  \\\n",
       "0                                      0                       0   \n",
       "1                                      0                       0   \n",
       "2                                      1                       0   \n",
       "3                                      0                       0   \n",
       "4                                      0                       0   \n",
       "...                                  ...                     ...   \n",
       "7390                                   0                       0   \n",
       "7391                                   0                       0   \n",
       "7392                                   0                       0   \n",
       "7393                                   0                       0   \n",
       "7394                                   0                       0   \n",
       "\n",
       "     alchemy_category_unknown alchemy_category_weather label_out id_out  \\\n",
       "0                           1                        0       1.0   3711   \n",
       "1                           1                        0       1.0   7222   \n",
       "2                           0                        0       1.0   3964   \n",
       "3                           0                        0       1.0   3697   \n",
       "4                           0                        0       1.0   4321   \n",
       "...                       ...                      ...       ...    ...   \n",
       "7390                        0                        0       NaN   3336   \n",
       "7391                        1                        0       NaN   2867   \n",
       "7392                        0                        0       NaN    738   \n",
       "7393                        1                        0       NaN   4124   \n",
       "7394                        0                        0       NaN   6056   \n",
       "\n",
       "                                      total_description  \n",
       "0     Join Polyvore follow latest stylepolyvore best...  \n",
       "1     Speed Air ManDavid BelleSpeed Air ManDavid Bel...  \n",
       "2     Chicken Gruyere RecipeChicken Gruyere one favo...  \n",
       "3     Ikan Bilis With Tempeh GroundnutsOh oh This re...  \n",
       "4     Malaysian recipeBarbecued Chicken Chow Siew Th...  \n",
       "...                                                 ...  \n",
       "7390  UnkonownUnkonownlanascooking 2011 07 08 roaste...  \n",
       "7391  Billy Chocolate Buttercream Martha Stewart Rec...  \n",
       "7392  Los Angeles Best Belly Dance Workout Swerve St...  \n",
       "7393  Buffalo Chicken Burgers recipe buffalo chicken...  \n",
       "7394  canarygirl com Sirloin Peppercorn Cream SauceS...  \n",
       "\n",
       "[7395 rows x 35 columns]"
      ]
     },
     "execution_count": 115,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "id": "2d8262b8",
   "metadata": {},
   "outputs": [],
   "source": [
    "main_df = pd.concat([main_df,df], axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "id": "12110fc5",
   "metadata": {},
   "outputs": [],
   "source": [
    "columns_to_be_deleted = [\"total_description\"]\n",
    "main_df.drop(axis=\"columns\", labels=columns_to_be_deleted, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "id": "9ede146a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1479"
      ]
     },
     "execution_count": 118,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df[\"label_out\"].isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "id": "3a01ed27",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>alchemy_category_score</th>\n",
       "      <th>avgLinkWordLength</th>\n",
       "      <th>AvglinkWithOneCommonWord</th>\n",
       "      <th>AvglinkWithTwoCommonWord</th>\n",
       "      <th>AvglinkWithThreeCommonWord</th>\n",
       "      <th>AvglinkWithFourCommonWord</th>\n",
       "      <th>redundancyMeasure</th>\n",
       "      <th>frameTagRatio</th>\n",
       "      <th>tagRatio</th>\n",
       "      <th>imageTagRatio</th>\n",
       "      <th>...</th>\n",
       "      <th>you</th>\n",
       "      <th>young</th>\n",
       "      <th>your</th>\n",
       "      <th>youre</th>\n",
       "      <th>youtube</th>\n",
       "      <th>yum</th>\n",
       "      <th>yummy</th>\n",
       "      <th>zest</th>\n",
       "      <th>zone</th>\n",
       "      <th>zucchini</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.869397</td>\n",
       "      <td>1.070441</td>\n",
       "      <td>2.850855</td>\n",
       "      <td>0.007905</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.589894</td>\n",
       "      <td>0.027778</td>\n",
       "      <td>0.203161</td>\n",
       "      <td>0.5529</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.869397</td>\n",
       "      <td>0.814292</td>\n",
       "      <td>3.129977</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.591068</td>\n",
       "      <td>0.015086</td>\n",
       "      <td>0.168189</td>\n",
       "      <td>0.5529</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.472093</td>\n",
       "      <td>1.10658</td>\n",
       "      <td>5.105662</td>\n",
       "      <td>0.250017</td>\n",
       "      <td>0.209826</td>\n",
       "      <td>0.184265</td>\n",
       "      <td>0.367008</td>\n",
       "      <td>0.033935</td>\n",
       "      <td>0.192652</td>\n",
       "      <td>0.126582</td>\n",
       "      <td>...</td>\n",
       "      <td>0.058072</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.045205</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.608077</td>\n",
       "      <td>0.980299</td>\n",
       "      <td>4.135385</td>\n",
       "      <td>0.181739</td>\n",
       "      <td>0.093401</td>\n",
       "      <td>0.06432</td>\n",
       "      <td>0.387063</td>\n",
       "      <td>0.03653</td>\n",
       "      <td>0.198</td>\n",
       "      <td>0.135593</td>\n",
       "      <td>...</td>\n",
       "      <td>0.019461</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.707464</td>\n",
       "      <td>0.457966</td>\n",
       "      <td>2.818948</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.256666</td>\n",
       "      <td>0.015152</td>\n",
       "      <td>0.241564</td>\n",
       "      <td>0.054217</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7390</th>\n",
       "      <td>1.976679</td>\n",
       "      <td>0.970779</td>\n",
       "      <td>3.925608</td>\n",
       "      <td>0.097638</td>\n",
       "      <td>0.033617</td>\n",
       "      <td>0.01695</td>\n",
       "      <td>0.591068</td>\n",
       "      <td>0.071719</td>\n",
       "      <td>0.237913</td>\n",
       "      <td>-0.2903</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7391</th>\n",
       "      <td>1.869397</td>\n",
       "      <td>1.174511</td>\n",
       "      <td>5.348241</td>\n",
       "      <td>0.208901</td>\n",
       "      <td>0.087011</td>\n",
       "      <td>0.020001</td>\n",
       "      <td>0.345951</td>\n",
       "      <td>0.056471</td>\n",
       "      <td>0.237229</td>\n",
       "      <td>0.024793</td>\n",
       "      <td>...</td>\n",
       "      <td>0.040439</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7392</th>\n",
       "      <td>2.50542</td>\n",
       "      <td>1.040202</td>\n",
       "      <td>4.538926</td>\n",
       "      <td>0.246225</td>\n",
       "      <td>0.054334</td>\n",
       "      <td>0.020101</td>\n",
       "      <td>0.436514</td>\n",
       "      <td>0.04129</td>\n",
       "      <td>0.21018</td>\n",
       "      <td>0.433526</td>\n",
       "      <td>...</td>\n",
       "      <td>0.023663</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7393</th>\n",
       "      <td>1.869397</td>\n",
       "      <td>1.009167</td>\n",
       "      <td>4.754497</td>\n",
       "      <td>0.204794</td>\n",
       "      <td>0.078643</td>\n",
       "      <td>0.018019</td>\n",
       "      <td>0.453249</td>\n",
       "      <td>0.070261</td>\n",
       "      <td>0.190801</td>\n",
       "      <td>0.5529</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7394</th>\n",
       "      <td>2.044258</td>\n",
       "      <td>1.105337</td>\n",
       "      <td>4.445692</td>\n",
       "      <td>0.196998</td>\n",
       "      <td>0.031749</td>\n",
       "      <td>0.027835</td>\n",
       "      <td>0.409636</td>\n",
       "      <td>0.012228</td>\n",
       "      <td>0.234657</td>\n",
       "      <td>0.078231</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>7395 rows × 3034 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     alchemy_category_score avgLinkWordLength AvglinkWithOneCommonWord  \\\n",
       "0                  1.869397          1.070441                 2.850855   \n",
       "1                  1.869397          0.814292                 3.129977   \n",
       "2                  1.472093           1.10658                 5.105662   \n",
       "3                  1.608077          0.980299                 4.135385   \n",
       "4                  1.707464          0.457966                 2.818948   \n",
       "...                     ...               ...                      ...   \n",
       "7390               1.976679          0.970779                 3.925608   \n",
       "7391               1.869397          1.174511                 5.348241   \n",
       "7392                2.50542          1.040202                 4.538926   \n",
       "7393               1.869397          1.009167                 4.754497   \n",
       "7394               2.044258          1.105337                 4.445692   \n",
       "\n",
       "     AvglinkWithTwoCommonWord AvglinkWithThreeCommonWord  \\\n",
       "0                    0.007905                        0.0   \n",
       "1                         0.0                        0.0   \n",
       "2                    0.250017                   0.209826   \n",
       "3                    0.181739                   0.093401   \n",
       "4                         0.0                        0.0   \n",
       "...                       ...                        ...   \n",
       "7390                 0.097638                   0.033617   \n",
       "7391                 0.208901                   0.087011   \n",
       "7392                 0.246225                   0.054334   \n",
       "7393                 0.204794                   0.078643   \n",
       "7394                 0.196998                   0.031749   \n",
       "\n",
       "     AvglinkWithFourCommonWord redundancyMeasure frameTagRatio  tagRatio  \\\n",
       "0                          0.0          0.589894      0.027778  0.203161   \n",
       "1                          0.0          0.591068      0.015086  0.168189   \n",
       "2                     0.184265          0.367008      0.033935  0.192652   \n",
       "3                      0.06432          0.387063       0.03653     0.198   \n",
       "4                          0.0          0.256666      0.015152  0.241564   \n",
       "...                        ...               ...           ...       ...   \n",
       "7390                   0.01695          0.591068      0.071719  0.237913   \n",
       "7391                  0.020001          0.345951      0.056471  0.237229   \n",
       "7392                  0.020101          0.436514       0.04129   0.21018   \n",
       "7393                  0.018019          0.453249      0.070261  0.190801   \n",
       "7394                  0.027835          0.409636      0.012228  0.234657   \n",
       "\n",
       "     imageTagRatio  ...       you young your youre youtube       yum yummy  \\\n",
       "0           0.5529  ...  0.000000   0.0  0.0   0.0     0.0  0.000000   0.0   \n",
       "1           0.5529  ...  0.000000   0.0  0.0   0.0     0.0  0.000000   0.0   \n",
       "2         0.126582  ...  0.058072   0.0  0.0   0.0     0.0  0.045205   0.0   \n",
       "3         0.135593  ...  0.019461   0.0  0.0   0.0     0.0  0.000000   0.0   \n",
       "4         0.054217  ...  0.000000   0.0  0.0   0.0     0.0  0.000000   0.0   \n",
       "...            ...  ...       ...   ...  ...   ...     ...       ...   ...   \n",
       "7390       -0.2903  ...  0.000000   0.0  0.0   0.0     0.0  0.000000   0.0   \n",
       "7391      0.024793  ...  0.040439   0.0  0.0   0.0     0.0  0.000000   0.0   \n",
       "7392      0.433526  ...  0.023663   0.0  0.0   0.0     0.0  0.000000   0.0   \n",
       "7393        0.5529  ...  0.000000   0.0  0.0   0.0     0.0  0.000000   0.0   \n",
       "7394      0.078231  ...  0.000000   0.0  0.0   0.0     0.0  0.000000   0.0   \n",
       "\n",
       "     zest zone zucchini  \n",
       "0     0.0  0.0      0.0  \n",
       "1     0.0  0.0      0.0  \n",
       "2     0.0  0.0      0.0  \n",
       "3     0.0  0.0      0.0  \n",
       "4     0.0  0.0      0.0  \n",
       "...   ...  ...      ...  \n",
       "7390  0.0  0.0      0.0  \n",
       "7391  0.0  0.0      0.0  \n",
       "7392  0.0  0.0      0.0  \n",
       "7393  0.0  0.0      0.0  \n",
       "7394  0.0  0.0      0.0  \n",
       "\n",
       "[7395 rows x 3034 columns]"
      ]
     },
     "execution_count": 119,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "id": "bd405e50",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>00</th>\n",
       "      <th>000</th>\n",
       "      <th>01</th>\n",
       "      <th>02</th>\n",
       "      <th>03</th>\n",
       "      <th>04</th>\n",
       "      <th>05</th>\n",
       "      <th>06</th>\n",
       "      <th>07</th>\n",
       "      <th>08</th>\n",
       "      <th>...</th>\n",
       "      <th>you</th>\n",
       "      <th>young</th>\n",
       "      <th>your</th>\n",
       "      <th>youre</th>\n",
       "      <th>youtube</th>\n",
       "      <th>yum</th>\n",
       "      <th>yummy</th>\n",
       "      <th>zest</th>\n",
       "      <th>zone</th>\n",
       "      <th>zucchini</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "      <td>7395.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.001946</td>\n",
       "      <td>0.003347</td>\n",
       "      <td>0.002842</td>\n",
       "      <td>0.003207</td>\n",
       "      <td>0.002825</td>\n",
       "      <td>0.003405</td>\n",
       "      <td>0.002950</td>\n",
       "      <td>0.002475</td>\n",
       "      <td>0.003070</td>\n",
       "      <td>0.003327</td>\n",
       "      <td>...</td>\n",
       "      <td>0.011420</td>\n",
       "      <td>0.001933</td>\n",
       "      <td>0.005036</td>\n",
       "      <td>0.001199</td>\n",
       "      <td>0.001614</td>\n",
       "      <td>0.000783</td>\n",
       "      <td>0.001299</td>\n",
       "      <td>0.001684</td>\n",
       "      <td>0.000661</td>\n",
       "      <td>0.002347</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.025414</td>\n",
       "      <td>0.022833</td>\n",
       "      <td>0.023438</td>\n",
       "      <td>0.025277</td>\n",
       "      <td>0.024095</td>\n",
       "      <td>0.029275</td>\n",
       "      <td>0.021842</td>\n",
       "      <td>0.019190</td>\n",
       "      <td>0.025241</td>\n",
       "      <td>0.026224</td>\n",
       "      <td>...</td>\n",
       "      <td>0.026450</td>\n",
       "      <td>0.013426</td>\n",
       "      <td>0.022905</td>\n",
       "      <td>0.014286</td>\n",
       "      <td>0.017827</td>\n",
       "      <td>0.009604</td>\n",
       "      <td>0.011086</td>\n",
       "      <td>0.014450</td>\n",
       "      <td>0.011699</td>\n",
       "      <td>0.033866</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.012880</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>0.792484</td>\n",
       "      <td>0.973723</td>\n",
       "      <td>0.763662</td>\n",
       "      <td>0.623520</td>\n",
       "      <td>0.840939</td>\n",
       "      <td>0.752708</td>\n",
       "      <td>0.531439</td>\n",
       "      <td>0.594531</td>\n",
       "      <td>0.560246</td>\n",
       "      <td>0.553521</td>\n",
       "      <td>...</td>\n",
       "      <td>0.532334</td>\n",
       "      <td>0.411285</td>\n",
       "      <td>0.541368</td>\n",
       "      <td>0.396205</td>\n",
       "      <td>0.680569</td>\n",
       "      <td>0.494492</td>\n",
       "      <td>0.285939</td>\n",
       "      <td>0.306863</td>\n",
       "      <td>0.758308</td>\n",
       "      <td>0.800716</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 3000 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                00          000           01           02           03  \\\n",
       "count  7395.000000  7395.000000  7395.000000  7395.000000  7395.000000   \n",
       "mean      0.001946     0.003347     0.002842     0.003207     0.002825   \n",
       "std       0.025414     0.022833     0.023438     0.025277     0.024095   \n",
       "min       0.000000     0.000000     0.000000     0.000000     0.000000   \n",
       "25%       0.000000     0.000000     0.000000     0.000000     0.000000   \n",
       "50%       0.000000     0.000000     0.000000     0.000000     0.000000   \n",
       "75%       0.000000     0.000000     0.000000     0.000000     0.000000   \n",
       "max       0.792484     0.973723     0.763662     0.623520     0.840939   \n",
       "\n",
       "                04           05           06           07           08  ...  \\\n",
       "count  7395.000000  7395.000000  7395.000000  7395.000000  7395.000000  ...   \n",
       "mean      0.003405     0.002950     0.002475     0.003070     0.003327  ...   \n",
       "std       0.029275     0.021842     0.019190     0.025241     0.026224  ...   \n",
       "min       0.000000     0.000000     0.000000     0.000000     0.000000  ...   \n",
       "25%       0.000000     0.000000     0.000000     0.000000     0.000000  ...   \n",
       "50%       0.000000     0.000000     0.000000     0.000000     0.000000  ...   \n",
       "75%       0.000000     0.000000     0.000000     0.000000     0.000000  ...   \n",
       "max       0.752708     0.531439     0.594531     0.560246     0.553521  ...   \n",
       "\n",
       "               you        young         your        youre      youtube  \\\n",
       "count  7395.000000  7395.000000  7395.000000  7395.000000  7395.000000   \n",
       "mean      0.011420     0.001933     0.005036     0.001199     0.001614   \n",
       "std       0.026450     0.013426     0.022905     0.014286     0.017827   \n",
       "min       0.000000     0.000000     0.000000     0.000000     0.000000   \n",
       "25%       0.000000     0.000000     0.000000     0.000000     0.000000   \n",
       "50%       0.000000     0.000000     0.000000     0.000000     0.000000   \n",
       "75%       0.012880     0.000000     0.000000     0.000000     0.000000   \n",
       "max       0.532334     0.411285     0.541368     0.396205     0.680569   \n",
       "\n",
       "               yum        yummy         zest         zone     zucchini  \n",
       "count  7395.000000  7395.000000  7395.000000  7395.000000  7395.000000  \n",
       "mean      0.000783     0.001299     0.001684     0.000661     0.002347  \n",
       "std       0.009604     0.011086     0.014450     0.011699     0.033866  \n",
       "min       0.000000     0.000000     0.000000     0.000000     0.000000  \n",
       "25%       0.000000     0.000000     0.000000     0.000000     0.000000  \n",
       "50%       0.000000     0.000000     0.000000     0.000000     0.000000  \n",
       "75%       0.000000     0.000000     0.000000     0.000000     0.000000  \n",
       "max       0.494492     0.285939     0.306863     0.758308     0.800716  \n",
       "\n",
       "[8 rows x 3000 columns]"
      ]
     },
     "execution_count": 120,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "bc28fef9",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "28620e4d",
   "metadata": {},
   "source": [
    "### Test - Train Split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "id": "00567aa0",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_df = main_df.loc[main_df[\"label_out\"].isna() == False,:]\n",
    "test_df = main_df.loc[main_df[\"label_out\"].isna(),:].drop(columns=\"label_out\")\n",
    "\n",
    "train_df = train_df.drop(columns=\"id_out\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "id": "208ea98d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>00</th>\n",
       "      <th>000</th>\n",
       "      <th>01</th>\n",
       "      <th>02</th>\n",
       "      <th>03</th>\n",
       "      <th>04</th>\n",
       "      <th>05</th>\n",
       "      <th>06</th>\n",
       "      <th>07</th>\n",
       "      <th>08</th>\n",
       "      <th>...</th>\n",
       "      <th>you</th>\n",
       "      <th>young</th>\n",
       "      <th>your</th>\n",
       "      <th>youre</th>\n",
       "      <th>youtube</th>\n",
       "      <th>yum</th>\n",
       "      <th>yummy</th>\n",
       "      <th>zest</th>\n",
       "      <th>zone</th>\n",
       "      <th>zucchini</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>5916.000000</td>\n",
       "      <td>5916.000000</td>\n",
       "      <td>5916.000000</td>\n",
       "      <td>5916.000000</td>\n",
       "      <td>5916.000000</td>\n",
       "      <td>5916.000000</td>\n",
       "      <td>5916.000000</td>\n",
       "      <td>5916.000000</td>\n",
       "      <td>5916.000000</td>\n",
       "      <td>5916.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>5916.000000</td>\n",
       "      <td>5916.000000</td>\n",
       "      <td>5916.000000</td>\n",
       "      <td>5916.000000</td>\n",
       "      <td>5916.000000</td>\n",
       "      <td>5916.000000</td>\n",
       "      <td>5916.000000</td>\n",
       "      <td>5916.000000</td>\n",
       "      <td>5916.000000</td>\n",
       "      <td>5916.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.001979</td>\n",
       "      <td>0.002991</td>\n",
       "      <td>0.002894</td>\n",
       "      <td>0.003105</td>\n",
       "      <td>0.002845</td>\n",
       "      <td>0.003581</td>\n",
       "      <td>0.003100</td>\n",
       "      <td>0.002516</td>\n",
       "      <td>0.003104</td>\n",
       "      <td>0.003215</td>\n",
       "      <td>...</td>\n",
       "      <td>0.011516</td>\n",
       "      <td>0.001947</td>\n",
       "      <td>0.005014</td>\n",
       "      <td>0.001361</td>\n",
       "      <td>0.001692</td>\n",
       "      <td>0.000812</td>\n",
       "      <td>0.001355</td>\n",
       "      <td>0.001775</td>\n",
       "      <td>0.000719</td>\n",
       "      <td>0.002340</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.025099</td>\n",
       "      <td>0.016532</td>\n",
       "      <td>0.023133</td>\n",
       "      <td>0.025206</td>\n",
       "      <td>0.023714</td>\n",
       "      <td>0.030247</td>\n",
       "      <td>0.021825</td>\n",
       "      <td>0.019547</td>\n",
       "      <td>0.025643</td>\n",
       "      <td>0.025805</td>\n",
       "      <td>...</td>\n",
       "      <td>0.026660</td>\n",
       "      <td>0.013810</td>\n",
       "      <td>0.022719</td>\n",
       "      <td>0.015662</td>\n",
       "      <td>0.018233</td>\n",
       "      <td>0.010139</td>\n",
       "      <td>0.011521</td>\n",
       "      <td>0.014793</td>\n",
       "      <td>0.012861</td>\n",
       "      <td>0.034100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.013127</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>0.792484</td>\n",
       "      <td>0.361182</td>\n",
       "      <td>0.763662</td>\n",
       "      <td>0.623520</td>\n",
       "      <td>0.840939</td>\n",
       "      <td>0.752708</td>\n",
       "      <td>0.429440</td>\n",
       "      <td>0.594531</td>\n",
       "      <td>0.560246</td>\n",
       "      <td>0.553521</td>\n",
       "      <td>...</td>\n",
       "      <td>0.532334</td>\n",
       "      <td>0.411285</td>\n",
       "      <td>0.392855</td>\n",
       "      <td>0.396205</td>\n",
       "      <td>0.680569</td>\n",
       "      <td>0.494492</td>\n",
       "      <td>0.285939</td>\n",
       "      <td>0.306863</td>\n",
       "      <td>0.758308</td>\n",
       "      <td>0.800716</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 3000 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                00          000           01           02           03  \\\n",
       "count  5916.000000  5916.000000  5916.000000  5916.000000  5916.000000   \n",
       "mean      0.001979     0.002991     0.002894     0.003105     0.002845   \n",
       "std       0.025099     0.016532     0.023133     0.025206     0.023714   \n",
       "min       0.000000     0.000000     0.000000     0.000000     0.000000   \n",
       "25%       0.000000     0.000000     0.000000     0.000000     0.000000   \n",
       "50%       0.000000     0.000000     0.000000     0.000000     0.000000   \n",
       "75%       0.000000     0.000000     0.000000     0.000000     0.000000   \n",
       "max       0.792484     0.361182     0.763662     0.623520     0.840939   \n",
       "\n",
       "                04           05           06           07           08  ...  \\\n",
       "count  5916.000000  5916.000000  5916.000000  5916.000000  5916.000000  ...   \n",
       "mean      0.003581     0.003100     0.002516     0.003104     0.003215  ...   \n",
       "std       0.030247     0.021825     0.019547     0.025643     0.025805  ...   \n",
       "min       0.000000     0.000000     0.000000     0.000000     0.000000  ...   \n",
       "25%       0.000000     0.000000     0.000000     0.000000     0.000000  ...   \n",
       "50%       0.000000     0.000000     0.000000     0.000000     0.000000  ...   \n",
       "75%       0.000000     0.000000     0.000000     0.000000     0.000000  ...   \n",
       "max       0.752708     0.429440     0.594531     0.560246     0.553521  ...   \n",
       "\n",
       "               you        young         your        youre      youtube  \\\n",
       "count  5916.000000  5916.000000  5916.000000  5916.000000  5916.000000   \n",
       "mean      0.011516     0.001947     0.005014     0.001361     0.001692   \n",
       "std       0.026660     0.013810     0.022719     0.015662     0.018233   \n",
       "min       0.000000     0.000000     0.000000     0.000000     0.000000   \n",
       "25%       0.000000     0.000000     0.000000     0.000000     0.000000   \n",
       "50%       0.000000     0.000000     0.000000     0.000000     0.000000   \n",
       "75%       0.013127     0.000000     0.000000     0.000000     0.000000   \n",
       "max       0.532334     0.411285     0.392855     0.396205     0.680569   \n",
       "\n",
       "               yum        yummy         zest         zone     zucchini  \n",
       "count  5916.000000  5916.000000  5916.000000  5916.000000  5916.000000  \n",
       "mean      0.000812     0.001355     0.001775     0.000719     0.002340  \n",
       "std       0.010139     0.011521     0.014793     0.012861     0.034100  \n",
       "min       0.000000     0.000000     0.000000     0.000000     0.000000  \n",
       "25%       0.000000     0.000000     0.000000     0.000000     0.000000  \n",
       "50%       0.000000     0.000000     0.000000     0.000000     0.000000  \n",
       "75%       0.000000     0.000000     0.000000     0.000000     0.000000  \n",
       "max       0.494492     0.285939     0.306863     0.758308     0.800716  \n",
       "\n",
       "[8 rows x 3000 columns]"
      ]
     },
     "execution_count": 122,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "id": "23fd57c3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "5916    7172\n",
       "5917    2687\n",
       "5918    6990\n",
       "5919    5639\n",
       "5920    2701\n",
       "        ... \n",
       "7390    3336\n",
       "7391    2867\n",
       "7392     738\n",
       "7393    4124\n",
       "7394    6056\n",
       "Name: id_out, Length: 1479, dtype: object"
      ]
     },
     "execution_count": 123,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_df[\"id_out\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "id": "0bca2789",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/sai/anaconda3/lib/python3.8/site-packages/seaborn/_decorators.py:36: FutureWarning: Pass the following variable as a keyword arg: x. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='label_out', ylabel='count'>"
      ]
     },
     "execution_count": 124,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.countplot(train_df[\"label_out\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "id": "a3a00293",
   "metadata": {},
   "outputs": [],
   "source": [
    "main_x = train_df.drop(columns=\"label_out\").to_numpy().astype(\"float\")\n",
    "main_y = train_df[\"label_out\"].to_numpy().astype(\"float\")\n",
    "\n",
    "train_x, test_x, train_y, test_y = train_test_split(main_x, main_y, train_size=0.6, stratify=main_y)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "12e8288f",
   "metadata": {},
   "source": [
    "# 2. Model Prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "id": "733798b6",
   "metadata": {},
   "outputs": [],
   "source": [
    "scaler = StandardScaler()\n",
    "train_x_scaled = scaler.fit_transform(train_x)\n",
    "\n",
    "test_x_scaled = scaler.transform(test_x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "id": "96a81756",
   "metadata": {},
   "outputs": [],
   "source": [
    "pca = PCA(n_components=300)\n",
    "train_x_pca = pca.fit_transform(train_x_scaled)\n",
    "test_x_pca = pca.transform(test_x_scaled)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dec595a7",
   "metadata": {},
   "source": [
    "## 2.1 Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "id": "90112e7e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression(C=0.5, class_weight='balanced', max_iter=10000)"
      ]
     },
     "execution_count": 128,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "logistic_regression_model = LogisticRegression(max_iter=10000, class_weight = \"balanced\", C=0.5)\n",
    "logistic_regression_model.fit(train_x_pca, train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "id": "54b6252c",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_yhat = logistic_regression_model.predict_proba(test_x_pca)[:,-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "id": "c65c1e1c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8540204941702549"
      ]
     },
     "execution_count": 130,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics.roc_auc_score(test_y, test_yhat)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a6f24177",
   "metadata": {},
   "source": [
    "## 2.2 Naive bayes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "id": "2d5ec10b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GaussianNB()"
      ]
     },
     "execution_count": 131,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gnb = GaussianNB()\n",
    "gnb.fit(train_x_pca, train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "id": "2695011e",
   "metadata": {},
   "outputs": [],
   "source": [
    "nb_test_yhat = gnb.predict_proba(test_x_pca)[:,-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "id": "a9d63e61",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7756695343637601"
      ]
     },
     "execution_count": 133,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics.roc_auc_score(test_y, nb_test_yhat)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0b012624",
   "metadata": {},
   "source": [
    "## 2.3 SVM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "id": "095686bb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(C=0.5, class_weight='balanced', probability=True)"
      ]
     },
     "execution_count": 134,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf = SVC(kernel = 'rbf',class_weight = \"balanced\",C = 0.5, probability = True)\n",
    "clf.fit(train_x_pca, train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "id": "f659bcd0",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_yhat = clf.predict_proba(test_x_pca)[:,-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "id": "8f9ce15c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8660924124680494"
      ]
     },
     "execution_count": 136,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics.roc_auc_score(test_y, test_yhat)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7efc5b50",
   "metadata": {},
   "source": [
    "### SVM with PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "id": "4f84e82e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(C=0.5, class_weight='balanced', probability=True)"
      ]
     },
     "execution_count": 137,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf_pca = SVC(kernel = 'rbf',class_weight = \"balanced\",C = 0.5, probability = True)\n",
    "clf_pca.fit(train_x_pca, train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "id": "3aeea3a8",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_yhat = clf_pca.predict_proba(test_x_pca)[:,-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "id": "7bd74059",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8660924124680494"
      ]
     },
     "execution_count": 139,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics.roc_auc_score(test_y, test_yhat)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "de44e005",
   "metadata": {},
   "source": [
    "## 2.4 ADABOOST\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "id": "7699f0b7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "AdaBoostClassifier(learning_rate=0.1, n_estimators=1000, random_state=0)"
      ]
     },
     "execution_count": 140,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf1 = AdaBoostClassifier(n_estimators=1000, random_state=0,learning_rate = 0.1)\n",
    "clf1.fit(train_x_pca, train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "id": "a48afed9",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_yhat_ada = clf1.predict_proba(test_x_pca)[:,-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "id": "1cdcc40f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8571690997775605"
      ]
     },
     "execution_count": 142,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics.roc_auc_score(test_y, test_yhat_ada)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "54fb56f0",
   "metadata": {},
   "source": [
    "## 2.5 Bagging"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "id": "fd403918",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "BaggingClassifier(base_estimator=SVC(), max_samples=30, n_estimators=1000,\n",
       "                  random_state=0)"
      ]
     },
     "execution_count": 143,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bagg = BaggingClassifier(base_estimator=SVC(),n_estimators=1000, random_state=0, max_samples = 30)\n",
    "bagg.fit(train_x_pca, train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "id": "ced8afe9",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_yhat_bag = bagg.predict_proba(test_x_pca)[:,-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "id": "89de8a0e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8530201596532458"
      ]
     },
     "execution_count": 145,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics.roc_auc_score(test_y, test_yhat_bag)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "767bc63f",
   "metadata": {},
   "source": [
    "### Mixing ADAboost and bagging"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "id": "816ca58b",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_yhat_new = np.maximum(test_yhat_bag,test_yhat_ada)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "id": "d595c08c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8536034200790261"
      ]
     },
     "execution_count": 147,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics.roc_auc_score(test_y, test_yhat_new)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a5dca30e",
   "metadata": {},
   "source": [
    "## 2.6 Decision Tree"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "id": "6ef6c8bc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeClassifier(class_weight='balanced', random_state=0)"
      ]
     },
     "execution_count": 148,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dt = DecisionTreeClassifier(class_weight = \"balanced\" ,random_state=0)\n",
    "dt.fit(train_x_pca, train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 149,
   "id": "71beada9",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_yhat_dt = dt.predict_proba(test_x_pca)[:,-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "id": "dba05c00",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.732360660799753"
      ]
     },
     "execution_count": 150,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics.roc_auc_score(test_y, test_yhat_dt)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ab7d0839",
   "metadata": {},
   "source": [
    "## 2.7 XGBoost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "id": "7f419117",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[12:09:22] WARNING: ../src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "XGBClassifier(base_score=0.5, booster='gbtree', colsample_bylevel=1,\n",
       "              colsample_bynode=1, colsample_bytree=1, enable_categorical=False,\n",
       "              gamma=0, gpu_id=-1, importance_type=None,\n",
       "              interaction_constraints='', learning_rate=0.1, max_delta_step=0,\n",
       "              max_depth=6, min_child_weight=1, missing=nan,\n",
       "              monotone_constraints='()', n_estimators=1000, n_jobs=8,\n",
       "              num_parallel_tree=1, predictor='auto', random_state=0,\n",
       "              reg_alpha=0, reg_lambda=1, scale_pos_weight=1, subsample=1,\n",
       "              tree_method='exact', use_label_encoder=False,\n",
       "              validate_parameters=1, verbosity=None)"
      ]
     },
     "execution_count": 153,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xgbc = xgb.XGBClassifier(n_estimators=1000, random_state=0,learning_rate = 0.1,use_label_encoder=False)\n",
    "xgbc.fit(train_x_pca, train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "id": "c3362045",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_yhat_xgbc = xgbc.predict_proba(test_x_pca)[:,-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 155,
   "id": "81fde3a2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.862372697693834"
      ]
     },
     "execution_count": 155,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics.roc_auc_score(test_y, test_yhat_xgbc)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f93070f8",
   "metadata": {},
   "source": [
    "## 2.8 Neural networks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 156,
   "id": "08fee5c4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "MLPClassifier(activation='logistic', alpha=1e-05,\n",
       "              hidden_layer_sizes=(40, 50, 60, 70, 60, 50, 40), max_iter=5000)"
      ]
     },
     "execution_count": 156,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nn = MLPClassifier(hidden_layer_sizes=(40,50,60,70,60,50,40), activation='logistic', solver='adam', max_iter=5000, alpha=1e-5)\n",
    "nn.fit(train_x_pca, train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 157,
   "id": "5f8e237a",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_yhat_clf = nn.predict_proba(test_x_pca)[:,-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 158,
   "id": "f80fd08e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8565379490962323"
      ]
     },
     "execution_count": 158,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics.roc_auc_score(test_y, test_yhat_clf)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "30aa5c03",
   "metadata": {},
   "source": [
    "## 3 Kaggle submission"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 159,
   "id": "3811533a",
   "metadata": {},
   "outputs": [],
   "source": [
    "main_x_scaled = scaler.fit_transform(main_x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 160,
   "id": "3c351cd2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1479, 3032)"
      ]
     },
     "execution_count": 160,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "kaggle_test_X = test_df.drop(columns=\"id_out\", inplace=False).to_numpy().astype(\"float\")\n",
    "\n",
    "kaggle_x_scaled = scaler.fit_transform(kaggle_test_X)\n",
    "\n",
    "kaggle_test_X.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 161,
   "id": "f2863c18",
   "metadata": {},
   "outputs": [],
   "source": [
    "main_x_pca = pca.fit_transform(main_x_scaled)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "id": "153f68ad",
   "metadata": {},
   "outputs": [],
   "source": [
    "kaggle_x_pca = pca.transform(kaggle_x_scaled)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 163,
   "id": "14572f6d",
   "metadata": {},
   "outputs": [],
   "source": [
    "testdf_new = test_df[\"id_out\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "id": "cdfb9c3b",
   "metadata": {},
   "outputs": [],
   "source": [
    "testdf_new = testdf_new.to_frame()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "id": "368a7ae9",
   "metadata": {},
   "outputs": [],
   "source": [
    "testdf_new[\"id\"] = test_df[\"id_out\"]\n",
    "testdf_new.drop(columns=\"id_out\",inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "88a62f22",
   "metadata": {},
   "source": [
    "### Logistic regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "id": "e8804505",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression(C=0.5, class_weight='balanced', max_iter=10000)"
      ]
     },
     "execution_count": 166,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "logistic_regression_model.fit(main_x_pca,main_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "id": "c5282576",
   "metadata": {},
   "outputs": [],
   "source": [
    "kaggle_test_yhat = logistic_regression_model.predict_proba(kaggle_x_pca)[:,-1]\n",
    "testdf_new[\"label\"] = kaggle_test_yhat \n",
    "to_submit = testdf_new[[\"id\", \"label\"]]\n",
    "to_submit.to_csv(\"phantom_Logistic_sub.csv\", index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "778797c1",
   "metadata": {},
   "source": [
    "### Naive Bayes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "id": "f02d3016",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GaussianNB()"
      ]
     },
     "execution_count": 168,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gnb.fit(main_x_pca,main_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 169,
   "id": "68cf8dda",
   "metadata": {},
   "outputs": [],
   "source": [
    "bayes_kaggle_test_yhat = gnb.predict_proba(kaggle_x_pca)[:,-1]\n",
    "testdf_new[\"label\"] = bayes_kaggle_test_yhat \n",
    "to_submit = testdf_new[[\"id\", \"label\"]]\n",
    "to_submit.to_csv(\"phantom_Bayes_sub.csv\", index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fede5c18",
   "metadata": {},
   "source": [
    "## svm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 170,
   "id": "b67c1f6b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(C=0.5, class_weight='balanced', probability=True)"
      ]
     },
     "execution_count": 170,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf.fit(main_x_pca,main_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 171,
   "id": "eaa397c3",
   "metadata": {},
   "outputs": [],
   "source": [
    "kaggle_test_yhat = clf.predict_proba(kaggle_x_pca)[:,-1]\n",
    "testdf_new[\"label\"] = kaggle_test_yhat \n",
    "to_submit = testdf_new[[\"id\", \"label\"]]\n",
    "to_submit.to_csv(\"phantom_svm_sub.csv\", index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cc74c6fc",
   "metadata": {},
   "source": [
    "### SVM with PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 172,
   "id": "3d088bcf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(C=0.5, class_weight='balanced', probability=True)"
      ]
     },
     "execution_count": 172,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf_pca.fit(main_x_pca,main_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 173,
   "id": "c79c0439",
   "metadata": {},
   "outputs": [],
   "source": [
    "kaggle_test_yhat = clf_pca.predict_proba(kaggle_x_pca)[:,-1]\n",
    "testdf_new[\"label\"] = kaggle_test_yhat \n",
    "to_submit = testdf_new[[\"id\", \"label\"]]\n",
    "to_submit.to_csv(\"phantom_svm_pca_sub.csv\", index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6ffadf86",
   "metadata": {},
   "source": [
    "## ADABOOST"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "id": "ec6f25c2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "AdaBoostClassifier(learning_rate=0.1, n_estimators=1000, random_state=0)"
      ]
     },
     "execution_count": 174,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf1.fit(main_x_pca,main_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 175,
   "id": "9940f5ff",
   "metadata": {},
   "outputs": [],
   "source": [
    "kaggle_test_yhat_ada = clf1.predict_proba(kaggle_x_pca)[:,-1]\n",
    "testdf_new[\"label\"] = kaggle_test_yhat_ada \n",
    "to_submit = testdf_new[[\"id\", \"label\"]]\n",
    "to_submit.to_csv(\"phantom_adaboost_sub.csv\", index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7c826d4d",
   "metadata": {},
   "source": [
    "## Bagging"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 176,
   "id": "1dbce8ff",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "BaggingClassifier(base_estimator=SVC(), max_samples=30, n_estimators=1000,\n",
       "                  random_state=0)"
      ]
     },
     "execution_count": 176,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bagg.fit(main_x_pca,main_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 177,
   "id": "24939a56",
   "metadata": {},
   "outputs": [],
   "source": [
    "kaggle_test_yhat_bagging = bagg.predict_proba(kaggle_x_pca)[:,-1]\n",
    "testdf_new[\"label\"] = kaggle_test_yhat_bagging \n",
    "to_submit = testdf_new[[\"id\", \"label\"]]\n",
    "to_submit.to_csv(\"phantom_bagging_sub.csv\", index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b7363e0e",
   "metadata": {},
   "source": [
    "## Mixing ADAboost and bagging"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 178,
   "id": "0fee311e",
   "metadata": {},
   "outputs": [],
   "source": [
    "kaggle_test_yhat_mix = (kaggle_test_yhat_bagging+kaggle_test_yhat_ada)/2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 179,
   "id": "029284b3",
   "metadata": {},
   "outputs": [],
   "source": [
    "testdf_new[\"label\"] = kaggle_test_yhat_mix \n",
    "to_submit = testdf_new[[\"id\", \"label\"]]\n",
    "to_submit.to_csv(\"phantom_mix_sub.csv\", index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f8b95f6e",
   "metadata": {},
   "source": [
    "## Decision Tree"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 180,
   "id": "e61907c6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeClassifier(class_weight='balanced', random_state=0)"
      ]
     },
     "execution_count": 180,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dt.fit(main_x_pca,main_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 181,
   "id": "e5fe31b4",
   "metadata": {},
   "outputs": [],
   "source": [
    "kaggle_test_yhat = dt.predict_proba(kaggle_x_pca)[:,-1]\n",
    "testdf_new[\"label\"] = kaggle_test_yhat \n",
    "to_submit = testdf_new[[\"id\", \"label\"]]\n",
    "to_submit.to_csv(\"phantom_decisiontree_sub.csv\", index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fe214ed1",
   "metadata": {},
   "source": [
    "## XGBoost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 182,
   "id": "55f2e694",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[12:19:56] WARNING: ../src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "XGBClassifier(base_score=0.5, booster='gbtree', colsample_bylevel=1,\n",
       "              colsample_bynode=1, colsample_bytree=1, enable_categorical=False,\n",
       "              gamma=0, gpu_id=-1, importance_type=None,\n",
       "              interaction_constraints='', learning_rate=0.1, max_delta_step=0,\n",
       "              max_depth=6, min_child_weight=1, missing=nan,\n",
       "              monotone_constraints='()', n_estimators=1000, n_jobs=8,\n",
       "              num_parallel_tree=1, predictor='auto', random_state=0,\n",
       "              reg_alpha=0, reg_lambda=1, scale_pos_weight=1, subsample=1,\n",
       "              tree_method='exact', use_label_encoder=False,\n",
       "              validate_parameters=1, verbosity=None)"
      ]
     },
     "execution_count": 182,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xgbc.fit(main_x_pca,main_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 183,
   "id": "5f176b6b",
   "metadata": {},
   "outputs": [],
   "source": [
    "kaggle_test_yhat = xgbc.predict_proba(kaggle_x_pca)[:,-1]\n",
    "testdf_new[\"label\"] = kaggle_test_yhat \n",
    "to_submit = testdf_new[[\"id\", \"label\"]]\n",
    "to_submit.to_csv(\"phantom_adaboost_sub.csv\", index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "226a3cbb",
   "metadata": {},
   "source": [
    "## Neural networks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "bfaf4779",
   "metadata": {},
   "outputs": [],
   "source": [
    "nn.fit(main_x_pca,main_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "33373f68",
   "metadata": {},
   "outputs": [],
   "source": [
    "kaggle_test_yhat = nn.predict_proba(kaggle_x_pca)[:,-1]\n",
    "testdf_new[\"label\"] = kaggle_test_yhat \n",
    "to_submit = testdf_new[[\"id\", \"label\"]]\n",
    "to_submit.to_csv(\"phantom_neural_network_sub.csv\", index=False)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}