3. Sentiment Analysis with Chi2 Square & PCA Dimension Reduction.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Sentiment analysis \n",
    "\n",
    "# Introduction\n",
    "Analyze & classify sentiment of text data, articles into positive or negative\n",
    "\n",
    "# Objective\n",
    "Sentiment analysis notebooks dives in very depth of various concepts, methods related to text analysis and understand the meaning of it semantically and/or syntactly. They are classified in the following five based notebooks based on different methods & tools used to analyze & classify text.\n",
    "\n",
    "1. Sentiment Analysis with Text Blob, Word Cloud, Count Vectorizer, N-Gram\n",
    "2. Sentiment Analysis using Doc2Vec, N-Gram & Phrase Modelling\n",
    "3. Sentiment Analysis with Chi2 Square & PCA Dimension Reduction\n",
    "4. Sentiment Analysis with Keras & Tensorflow\n",
    "5. Sentiment Analysis with Keras & Tensorflow using Doc2Vec, Pretrained GloVe"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Tres\n",
    "\n",
    "## 3. Sentiment Analysis with Text Blob, Word Cloud, Count Vectorizer, N-Gram, Zif2 law, Lexical Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Basic import\n",
    "\n",
    "import re\n",
    "import pandas as pd  \n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "plt.style.use('fivethirtyeight')\n",
    "\n",
    "%matplotlib inline\n",
    "%config InlineBackend.figure_format = 'retina'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tqdm import tqdm\n",
    "tqdm.pandas(desc=\"progress-bar\")\n",
    "\n",
    "from gensim.models import Doc2Vec\n",
    "from gensim.models.doc2vec import LabeledSentence\n",
    "from gensim.models.phrases import Phrases, Phraser"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from textblob import TextBlob\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.metrics import classification_report, confusion_matrix\n",
    "\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.pipeline import Pipeline\n",
    "from datetime import datetime\n",
    "\n",
    "import multiprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 1600000 entries, 0 to 1599999\n",
      "Data columns (total 3 columns):\n",
      "sentiment        1600000 non-null int64\n",
      "text             1600000 non-null object\n",
      "pre_clean_len    1600000 non-null int64\n",
      "dtypes: int64(2), object(1)\n",
      "memory usage: 48.8+ MB\n"
     ]
    },
    {
     "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>sentiment</th>\n",
       "      <th>text</th>\n",
       "      <th>pre_clean_len</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>awww that bummer you shoulda got david carr of...</td>\n",
       "      <td>115</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>is upset that he can not update his facebook b...</td>\n",
       "      <td>111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>dived many times for the ball managed to save ...</td>\n",
       "      <td>89</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>my whole body feels itchy and like its on fire</td>\n",
       "      <td>47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>no it not behaving at all mad why am here beca...</td>\n",
       "      <td>111</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   sentiment                                               text  pre_clean_len\n",
       "0          0  awww that bummer you shoulda got david carr of...            115\n",
       "1          0  is upset that he can not update his facebook b...            111\n",
       "2          0  dived many times for the ball managed to save ...             89\n",
       "3          0     my whole body feels itchy and like its on fire             47\n",
       "4          0  no it not behaving at all mad why am here beca...            111"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Read TF dataframe\n",
    "\n",
    "df = pd.read_hdf('./data/redstone.hdf')\n",
    "df.info()\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 1600000 entries, 0 to 1599999\n",
      "Data columns (total 3 columns):\n",
      "sentiment        1600000 non-null int64\n",
      "text             1600000 non-null object\n",
      "pre_clean_len    1600000 non-null int64\n",
      "dtypes: int64(2), object(1)\n",
      "memory usage: 36.6+ MB\n"
     ]
    }
   ],
   "source": [
    "# Santitizing dataframe\n",
    "\n",
    "df.dropna(inplace=True)\n",
    "df.reset_index(drop=True,inplace=True)\n",
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/indiano/miniconda3/envs/tf/lib/python3.5/site-packages/sklearn/cross_validation.py:41: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.\n",
      "  \"This module will be removed in 0.20.\", DeprecationWarning)\n"
     ]
    }
   ],
   "source": [
    "from sklearn import utils\n",
    "from sklearn.cross_validation import train_test_split\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "\n",
    "train = df.text\n",
    "label = df.sentiment\n",
    "SEED = 21\n",
    "\n",
    "# Splitting data into train, test & validation sets\n",
    "x_train, x_val_test, y_train, y_val_test = train_test_split(train, label, test_size=.02, random_state=SEED)\n",
    "\n",
    "x_val, x_test, y_val, y_test = train_test_split(x_val_test, y_val_test, test_size=.5, random_state=SEED)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train set entries.\n",
      "50.00% Negative Entries\n",
      "50.00% Positive Entries\n",
      "\n",
      "Validation set entries.\n",
      "50.01% Negative Entries\n",
      "49.99% Positive Entries\n",
      "\n",
      "Test set entries.\n",
      "50.21% Negative Entries\n",
      "49.79% Positive Entries\n"
     ]
    }
   ],
   "source": [
    "# Quantifying the positive & negative sentiments in the dataset\n",
    "\n",
    "from collections import Counter\n",
    "\n",
    "counter = Counter(y_train)\n",
    "print('Train set entries.')\n",
    "for key in counter:\n",
    "    if key == 0:\n",
    "        print('{:.2f}% Negative Entries'.format( (counter[key]/len(y_train))*100 ))\n",
    "    elif key == 1:\n",
    "        print('{:.2f}% Positive Entries'.format( (counter[key]/len(y_train))*100 ))\n",
    "        \n",
    "counter = Counter(y_val)\n",
    "print('\\nValidation set entries.')\n",
    "for key in counter:\n",
    "    if key == 0:\n",
    "        print('{:.2f}% Negative Entries'.format( (counter[key]/len(y_val))*100 ))\n",
    "    elif key == 1:\n",
    "        print('{:.2f}% Positive Entries'.format( (counter[key]/len(y_val))*100 ))\n",
    "\n",
    "counter = Counter(y_test)\n",
    "print('\\nTest set entries.')\n",
    "for key in counter:\n",
    "    if key == 0:\n",
    "        print('{:.2f}% Negative Entries'.format( (counter[key]/len(y_test))*100 ))\n",
    "    elif key == 1:\n",
    "        print('{:.2f}% Positive Entries'.format( (counter[key]/len(y_test))*100 ))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculate accuracy & summmary of different set of features\n",
    "\n",
    "def accuracy_features(pipeline, x_train, y_train, x_test, y_test):\n",
    "    \n",
    "    counter = Counter(y_test)\n",
    "\n",
    "    if (counter[0] / (len(y_test)*1.)) > 0.5:\n",
    "        baseline_accuracy = counter[0] / (len(y_test)*1.)\n",
    "    else:\n",
    "        baseline_accuracy = 1. - (counter[0] / (len(y_test)*1.))\n",
    "   \n",
    "    # Timer starts\n",
    "    timer = datetime.now()\n",
    "    \n",
    "    model = pipeline.fit(x_train, y_train)\n",
    "    y_pred = model.predict(x_test)\n",
    "    \n",
    "    elapsed_time = datetime.now() - timer\n",
    "    # Timer stops\n",
    "\n",
    "    accuracy = accuracy_score(y_test, y_pred)\n",
    "    \n",
    "    \n",
    "    print('Baseline accuracy: {:.2f}%'.format(baseline_accuracy*100))\n",
    "    print('Accuracy score: {:.2f}%'.format(accuracy*100))\n",
    "    \n",
    "    if(accuracy > baseline_accuracy):\n",
    "        print('\\nModel accuracy:{:.2f}% - Baseline accuracy:{:.2f}%: Increase of {:.2f}%'.format(accuracy*100, baseline_accuracy*100, (accuracy-baseline_accuracy)*100))\n",
    "    else:\n",
    "        print('Model accuracy:{:.2f}% - Baseline accuracy:{:.2f}%: Decrease of {:.2f}%'.format(accuracy*100, baseline_accuracy*100, (accuracy-baseline_accuracy)*100))\n",
    "    \n",
    "    print('Overall Train and Prediction time: {:.2f}s'.format(elapsed_time.total_seconds()))\n",
    "    print('-'*89)\n",
    "          \n",
    "    return accuracy, elapsed_time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Feature extraction, Iteratively\n",
    "\n",
    "# Count Vectorizer, Logistic Regression model\n",
    "cvec = CountVectorizer()\n",
    "lr = LogisticRegression()\n",
    "\n",
    "n_features = np.arange(10000, 100001, 10000)\n",
    "\n",
    "def classical_feature_extraction(vectorizer=cvec, n_features=n_features, stop_words=None, ngram_range=(1, 1), classifier=lr):\n",
    "    result = []\n",
    "    print(classifier)\n",
    "\n",
    "    for features in n_features:\n",
    "        # Setting vectorizer parameters & creating pipeline\n",
    "        cvec.set_params(stop_words=stop_words, max_features=features, ngram_range=ngram_range)\n",
    "        pipeline = Pipeline([\n",
    "            ('vectorizer', cvec),\n",
    "            ('classifier', classifier)\n",
    "        ])\n",
    "        print(\"\\nValidation result for {} features\".format(features))\n",
    "        \n",
    "        # Calculate accuracy & summmary\n",
    "        features_accuracy, features_time = accuracy_features(pipeline, x_train, y_train, x_val, y_val)\n",
    "        result.append((features, features_accuracy, features_time))\n",
    "        \n",
    "    return result"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### TFIDF Vectorizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "tvec = TfidfVectorizer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Result for TRIGRAM With STOP words using tfidf\n",
      "\n",
      "LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,\n",
      "          intercept_scaling=1, max_iter=100, multi_class='ovr', n_jobs=1,\n",
      "          penalty='l2', random_state=None, solver='liblinear', tol=0.0001,\n",
      "          verbose=0, warm_start=False)\n",
      "\n",
      "Validation result for 10000 features\n",
      "Baseline accuracy: 50.01%\n",
      "Accuracy score: 80.29%\n",
      "\n",
      "Model accuracy:80.29% - Baseline accuracy:50.01%: Increase of 30.28%\n",
      "Overall Train and Prediction time: 214.64s\n",
      "-----------------------------------------------------------------------------------------\n",
      "\n",
      "Validation result for 20000 features\n",
      "Baseline accuracy: 50.01%\n",
      "Accuracy score: 81.25%\n",
      "\n",
      "Model accuracy:81.25% - Baseline accuracy:50.01%: Increase of 31.24%\n",
      "Overall Train and Prediction time: 261.76s\n",
      "-----------------------------------------------------------------------------------------\n",
      "\n",
      "Validation result for 30000 features\n",
      "Baseline accuracy: 50.01%\n",
      "Accuracy score: 81.90%\n",
      "\n",
      "Model accuracy:81.90% - Baseline accuracy:50.01%: Increase of 31.89%\n",
      "Overall Train and Prediction time: 269.03s\n",
      "-----------------------------------------------------------------------------------------\n",
      "\n",
      "Validation result for 40000 features\n",
      "Baseline accuracy: 50.01%\n",
      "Accuracy score: 82.17%\n",
      "\n",
      "Model accuracy:82.17% - Baseline accuracy:50.01%: Increase of 32.16%\n",
      "Overall Train and Prediction time: 299.71s\n",
      "-----------------------------------------------------------------------------------------\n",
      "\n",
      "Validation result for 50000 features\n",
      "Baseline accuracy: 50.01%\n",
      "Accuracy score: 82.32%\n",
      "\n",
      "Model accuracy:82.32% - Baseline accuracy:50.01%: Increase of 32.31%\n",
      "Overall Train and Prediction time: 305.94s\n",
      "-----------------------------------------------------------------------------------------\n",
      "\n",
      "Validation result for 60000 features\n",
      "Baseline accuracy: 50.01%\n",
      "Accuracy score: 82.29%\n",
      "\n",
      "Model accuracy:82.29% - Baseline accuracy:50.01%: Increase of 32.27%\n",
      "Overall Train and Prediction time: 283.57s\n",
      "-----------------------------------------------------------------------------------------\n",
      "\n",
      "Validation result for 70000 features\n",
      "Baseline accuracy: 50.01%\n",
      "Accuracy score: 82.24%\n",
      "\n",
      "Model accuracy:82.24% - Baseline accuracy:50.01%: Increase of 32.23%\n",
      "Overall Train and Prediction time: 372.79s\n",
      "-----------------------------------------------------------------------------------------\n",
      "\n",
      "Validation result for 80000 features\n",
      "Baseline accuracy: 50.01%\n",
      "Accuracy score: 82.16%\n",
      "\n",
      "Model accuracy:82.16% - Baseline accuracy:50.01%: Increase of 32.14%\n",
      "Overall Train and Prediction time: 323.39s\n",
      "-----------------------------------------------------------------------------------------\n",
      "\n",
      "Validation result for 90000 features\n",
      "Baseline accuracy: 50.01%\n",
      "Accuracy score: 82.13%\n",
      "\n",
      "Model accuracy:82.13% - Baseline accuracy:50.01%: Increase of 32.12%\n",
      "Overall Train and Prediction time: 361.10s\n",
      "-----------------------------------------------------------------------------------------\n",
      "\n",
      "Validation result for 100000 features\n",
      "Baseline accuracy: 50.01%\n",
      "Accuracy score: 82.06%\n",
      "\n",
      "Model accuracy:82.06% - Baseline accuracy:50.01%: Increase of 32.05%\n",
      "Overall Train and Prediction time: 391.38s\n",
      "-----------------------------------------------------------------------------------------\n",
      "CPU times: user 2h 19min 44s, sys: 1min 50s, total: 2h 21min 34s\n",
      "Wall time: 51min 24s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "\n",
    "print('Result for TRIGRAM With STOP words using tfidf\\n')\n",
    "features = classical_feature_extraction(vectorizer=tvec, ngram_range=(1, 3))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Trigram Features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Dataframes feature extraction results for without stop words \n",
    "\n",
    "fdf = pd.DataFrame(features, columns=['features', 'features_accuracy', 'features_time'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "fdf.to_hdf('./data/tf_trigram_features.hdf', 'tf_trigram_features')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 10 entries, 0 to 9\n",
      "Data columns (total 3 columns):\n",
      "features             10 non-null int64\n",
      "features_accuracy    10 non-null float64\n",
      "features_time        10 non-null timedelta64[ns]\n",
      "dtypes: float64(1), int64(1), timedelta64[ns](1)\n",
      "memory usage: 320.0 bytes\n"
     ]
    }
   ],
   "source": [
    "fdf = pd.read_hdf('./data/tf_trigram_features.hdf')\n",
    "fdf.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Vectorize train, validation sets using above dbow_ug_model (Document Bag Of Words Unigram Model)\n",
    "\n",
    "def vectorize_concate(model1, model2, corpus, size):\n",
    "    # Numpy zeros initialization\n",
    "    vectors = np.zeros((len(corpus), size))\n",
    "    \n",
    "    for idx, count in zip(corpus.index, range(len(corpus.index))):\n",
    "        prefix = 'all_' + str(idx)\n",
    "        # Appending document vectors\n",
    "        vectors[count] = np.append(model1.docvecs[prefix], model2.docvecs[prefix])\n",
    "\n",
    "    return vectors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load the dbow_ug_model and delete temporary training data\n",
    "\n",
    "dbow_ug_model = Doc2Vec.load('./data/dbow_ug_model.doc2vec')\n",
    "dbow_ug_model.delete_temporary_training_data(keep_doctags_vectors=True, keep_inference=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load the dmm_ug_model and delete temporary training data\n",
    "\n",
    "dmm_tg_model = Doc2Vec.load('./data/dmm_tg_model.doc2vec')\n",
    "dmm_tg_model.delete_temporary_training_data(keep_doctags_vectors=True, keep_inference=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Vectorize & concate document vectors of train, validation sets obtained from Distributed Bag Of Words & Distributed Memory \n",
    "\n",
    "train_vecs_dbow_dmm = vectorize_concate(dbow_ug_model, dmm_tg_model, x_train, 200)\n",
    "val_vecs_dbow_dmm = vectorize_concate(dbow_ug_model, dmm_tg_model, x_val, 200)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Dimensionality Reduction¶"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "After tring three different methods (Count Vectorizer, TFIDF Vectorizer, DOC2Vec) for feature extraction, the best result on the validation set using classical Logistic Regression classifier (L2 regularization, regularization strength: 1.0) for each of them are as follow."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Populate table with Feature Extractors & it's Accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "mydata = [['Count Vectorizer', 80000, [1,2,3], 'Yes', '82.44%'], \n",
    "          ['TFIDF Vectorizer', 100000, [1,2,3], 'Yes', '82.92%'],\n",
    "          ['DOC2VEC', 200, 'Unigram DBOW + Trigram DMM', 'Yes', '75.76%'],\n",
    "          ['Custom Lexicon Classifier', '-', 'Decision Threshold set to 0.56', '-', '75.96%']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table>\n",
       "<thead>\n",
       "<tr><th>Feature Extractor        </th><th>No. Of Features  </th><th>N-Gram                        </th><th>Stop Words  </th><th>Accuracy  </th></tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "<tr><td>Count Vectorizer         </td><td>80000            </td><td>[1, 2, 3]                     </td><td>Yes         </td><td>82.44%    </td></tr>\n",
       "<tr><td>TFIDF Vectorizer         </td><td>100000           </td><td>[1, 2, 3]                     </td><td>Yes         </td><td>82.92%    </td></tr>\n",
       "<tr><td>DOC2VEC                  </td><td>200              </td><td>Unigram DBOW + Trigram DMM    </td><td>Yes         </td><td>75.76%    </td></tr>\n",
       "<tr><td>Custom Lexicon Classifier</td><td>-                </td><td>Decision Threshold set to 0.56</td><td>-           </td><td>75.96%    </td></tr>\n",
       "</tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from tabulate import tabulate\n",
    "from IPython.display import HTML\n",
    "\n",
    "display(HTML(tabulate(mydata, headers= ['Feature Extractor', 'No. Of Features', 'N-Gram', 'Stop Words', 'Accuracy'], floatfmt='.6f', tablefmt='html')))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Apart from the lexical approach, the number of features needed for each model seems quite big. Hence It would be interesting to see how model behaves after reducing dimensions (TFIDF Vectorizer & DOC2VEC) and it would also be easy to do hyperparameter tuning on the reduced features.\n",
    "\n",
    "The no. of dimensions from Doc2Vec model is 200 (dense matrices with all real numbers), sounds quite small as compared to 100,000 features (sparse matrices with lots of zeros) from TFIDF Vectorizer but the computation time for Doc2Vec is not so small."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### TFIDF Dimensionality reduction with chi-squared feature selection.\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Chi2 Feature Selection"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In Sklearn, there are three methods you can use for feature selection with sparse matrices such as Tfidf vectors or count vectors. \n",
    "\n",
    "Feature selection in Sklearn\n",
    "\n",
    "1.1) Removing features with low variance : VarianceThreshold\n",
    "\n",
    "1.2) Univariate feature selection \n",
    "\n",
    "Scoring functions           \n",
    "For regression: f_regression, mutual_info_regression      \n",
    "For classification: chi2, f_classif, mutual_info_classif\n",
    "\n",
    "Feature selection with sparse data\n",
    "\n",
    "If you use sparse data (i.e. data represented as sparse matrices), chi2, mutual_info_regression, mutual_info_classif will deal with the data without making it dense.\n",
    "\n",
    "1.3) Recursive feature elimination\n",
    "\n",
    "1.4) Feature selection using SelectFromModel\n",
    "    \n",
    "    1.4.1.1) L1-based feature selection\n",
    "    1.4.1.2) L1-recovery and compressive sensing\n",
    "\n",
    "    1.4.2) Tree-based feature selection\n",
    "\n",
    "1.5) Feature selection as part of a pipeline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In my case, I have 1.5 million tweets and want to reduce dimensions from 100,000 features, hence transforming tweets into dense matrices is not an option as it will not fit on my RAM.\n",
    "\n",
    "The Chi-squared statistic measures the lack of independence between a feature (in this case, one term within a tweet) and class (whether the tweets are positive or negative).\n",
    "\n",
    "You can first think of the relationship between a term within a tweet and it's class in terms of a contingency table. The contingency table is just a fancy word for a table that displays frequency distribution."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Chi2 Feature selection Example with beloved dogs\n",
    "\n",
    "Let's say we have three sentences in our tweet corpus as below.\n",
    "\n",
    "1) I love dogs              \n",
    "2) I hate dogs                  \n",
    "3) I love dogs and cooking"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Defining the sentiment class of each of the above tweet "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "mydata = [[\"'dogs' occur\", '2 (A)', '0 (B)'], \n",
    "          [\"'dogs' doesn't occur\", '2 (A)', '0 (B)']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table>\n",
       "<thead>\n",
       "<tr><th>                    </th><th>Positive Class  </th><th>Negative Class  </th></tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "<tr><td>'dogs' occur        </td><td>2 (A)           </td><td>0 (B)           </td></tr>\n",
       "<tr><td>'dogs' doesn't occur</td><td>2 (A)           </td><td>0 (B)           </td></tr>\n",
       "</tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "HTML(tabulate(mydata, headers= ['', 'Positive Class', 'Negative Class'], tablefmt='html'))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's look for a specific term, in this case beloved 'dogs' as 't' and class as 'c' (Positive or Negative). Using contingency table, where A is the number of times 't' occurs and 'c' is positive; B is the number of times 't' occurs and 'c' is negative; C is the number of times 't' doesn't occur and 'c' is positive; finally D is the number of times 't' doesn't occur and 'c' is negative. \n",
    "\n",
    "Now we are ready to calculate Chi-squared statistic.\n",
    "\n",
    "$${\\chi}^2(t,c) = \\frac {{N}\\times{(AD-CB)}^2}{(A+C)\\times(B+D)\\times(A+B)\\times(C+D)}$$\n",
    "\n",
    "Where N is the total number of samples.\n",
    "\n",
    "Chi2 score for term \"dogs\" is 3. As Chi2 square measures the lack of independence between a feature and it's class, hence if a feature has high chi-squared score compared to other features then that feature exhibits important predictive power for the class."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's first transform the train data into TFIDF vectors of 100,000 features and reduce dimensions using chi2square."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.feature_selection import SelectKBest, chi2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "tvec = TfidfVectorizer(max_features=100000, ngram_range=(1, 3))\n",
    "\n",
    "tf_train, tf_val = tvec.fit_transform(x_train), tvec.transform(x_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Fitting Chi2\n",
    "\n",
    "chi2score = chi2(tf_train, y_train)[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "100000"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "chi2score.size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Dumping chi2score\n",
    "\n",
    "import pickle\n",
    "\n",
    "with open('./data/chi2score.pkl', 'wb') as file:\n",
    "    pickle.dump(chi2score, file)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0,0.5,'Features')"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 591,
       "width": 1024
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot Important chi2square features after reducing dimesions\n",
    "\n",
    "plt.figure(figsize=(15,10))\n",
    "\n",
    "# Zip features names alongwith their scores & sort by chi2 score\n",
    "features_chi2scores = zip(tvec.get_feature_names(), chi2score)\n",
    "features_chi2scores_sorted = sorted(features_chi2scores, key=lambda x:x[1])\n",
    "\n",
    "# Top 20 Chi2 features\n",
    "topchi2 = list(zip(*features_chi2scores_sorted[-20:]))\n",
    "\n",
    "# Feature name & total length of the features to plot a horizontal barplot\n",
    "feature_names = topchi2[0]\n",
    "feature_score = topchi2[1]\n",
    "features_len = range(len(feature_score))\n",
    "\n",
    "plt.barh(features_len, feature_score, align='center', alpha=0.2)\n",
    "plt.plot(feature_score, features_len, '-o', markersize=5, alpha=0.8)\n",
    "\n",
    "plt.yticks(features_len, feature_names)\n",
    "\n",
    "plt.xlabel('Chi2 $\\chi^2$')\n",
    "plt.ylabel('Features')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The most useful feature selected by chi2 is the word \"thanks\" and I assume that this came mostly from positive tweets. The second most useful features it chose is the word \"sad\" and this time I guess it came from negative tweets. \n",
    "\n",
    "If you consider how chi2 is calculated, it will not only score highly on predictive terms of positive class but also score highly on predictive terms of negative class.\n",
    "\n",
    "Since now we have an idea of how the chi2 feature selection works, let's reduce the dimensions to different number of features, and also check the accuracy on the validation set."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "One more thing. Tfidf vectorizer can limit the number of features in the parameter setting just before fit and transform method. \n",
    "\n",
    "I want to compare the validation accuracy on the same number of features as follow.\n",
    "1) when the number of features has been limited during tfidf vectorizer initialization.\n",
    "2) when the number of features has been reduced from 100,000 features using chi2 statistic & tfidf vectorizer initialized with max 100000 features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Evaluation done for 10000 features using Chi2 feature selection method\n",
      "Evaluation done for 20000 features using Chi2 feature selection method\n",
      "Evaluation done for 30000 features using Chi2 feature selection method\n",
      "Evaluation done for 40000 features using Chi2 feature selection method\n",
      "Evaluation done for 50000 features using Chi2 feature selection method\n",
      "Evaluation done for 60000 features using Chi2 feature selection method\n",
      "Evaluation done for 70000 features using Chi2 feature selection method\n",
      "Evaluation done for 80000 features using Chi2 feature selection method\n",
      "Evaluation done for 90000 features using Chi2 feature selection method\n"
     ]
    }
   ],
   "source": [
    "# Dimension reduction & testing on valdiation set\n",
    "\n",
    "ch2_result = []\n",
    "\n",
    "for features in np.arange(10000, 100000, 10000):\n",
    "    # Select best features\n",
    "    ch2 = SelectKBest(chi2, k=features)\n",
    "    \n",
    "    x_train_chi2_selected = ch2.fit_transform(tf_train, y_train)\n",
    "    x_val_chi2_selected = ch2.transform(tf_val)\n",
    "                      \n",
    "    # Fitting Logistic Regression classical model\n",
    "    clf = LogisticRegression()\n",
    "    clf.fit(x_train_chi2_selected, y_train)\n",
    "    \n",
    "    # Testing on the validation set\n",
    "    score = clf.score(x_val_chi2_selected, y_val)\n",
    "                      \n",
    "    ch2_result.append(score)\n",
    "                      \n",
    "    print('Evaluation done for {} features using Chi2 feature selection method'.format(features))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f7c9be07fd0>"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 604,
       "width": 1253
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot Important chi2square features after reducing dimesions\n",
    "\n",
    "plt.figure(figsize=(15,10))\n",
    "\n",
    "plt.plot(fdf.features, fdf.features_accuracy, label='Trigram TFIDF Vectorizer', color='royalblue')\n",
    "plt.plot(np.arange(10000, 100000, 10000), ch2_result, label='Uniigram+Trigram TFIDF dimesion reduction (from 100,000 features) using Chi2 square', linestyle=':', color='orangered')\n",
    "\n",
    "plt.xlabel(\"Number of features\")\n",
    "plt.ylabel(\"Validation set accuracy\")\n",
    "\n",
    "plt.title(\"TFIDF vectorizer: Features limited during tfidf vectorizer initialization VS tfidf vectorizer initialized with max 100000 features & dimensionally reduced with chi2 afterwards\")\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Principal Component Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "PCA transforms the coordinate system in such a way that the resultant principal component axes describe the data in more concise, informative way."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![title](images/pca.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The first Principal Component explains the most variance in the data. In figure B, we see that by drawing the line \"Component 1\", it is able to retain a lot of information about the dispersed data points.\n",
    "\n",
    "The second Principal Component \"Component 2\" explains the second most variance in the data.\n",
    "\n",
    "The next step is to transform the original data onto our new found pca 2D axes. The final result in figure C gives us a pretty good idea of how the data is redistributed on 2D space even after discarding the third dimension.\n",
    "\n",
    "Eigenvectors & Eigenvalues are used while implementing above transformations on the data. \n",
    "\n",
    "An Eigenvector specifies a direction relative to the original coordinate space, while Eigenvalues indicate the magnitude or amount of variance in the direction of its corresponding Eigenvector."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Blogs to deep dive futher into the concept of PCA\n",
    "\n",
    "- http://mengnote.blogspot.co.uk/2013/05/an-intuitive-explanation-of-pca.html\n",
    "\n",
    "- A blog post by Victor Powell: \"Principal Component Analysis\" (You can manipulate and play around with data points in 2d or 3d that will help you enormously to have an intuitive understanding of what PCA does) http://setosa.io/ev/principal-component-analysis/\n",
    "\n",
    "- A blog post by George Dallas: \"Principal Component Analysis 4 Dummies: Eigenvectors, Eigenvalues and Dimension Reduction\" (he has successfully explained the concept without a single math formula, unbelievable) https://georgemdallas.wordpress.com/2013/10/30/principal-component-analysis-4-dummies-eigenvectors-eigenvalues-and-dimension-reduction/"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, let's try to reduce dimensions of doc2vec vectors with PCA. We can also plot the result on a graph and see if it make sense to reduce the number of features into a smaller set of principal components and how much a given number of principal components with higher variance, can explain about the original features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Scaling inputs\n",
    "\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "scaler = StandardScaler()\n",
    "\n",
    "# Scaling train and val sets\n",
    "train_vecs_dbow_dmm_scaled = scaler.fit_transform(train_vecs_dbow_dmm)\n",
    "val_vecs_dbow_dmm_scaled = scaler.fit_transform(val_vecs_dbow_dmm)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# PCA Decomposition\n",
    "\n",
    "from sklearn.decomposition import PCA\n",
    "\n",
    "pca = PCA().fit(train_vecs_dbow_dmm_scaled)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 386,
       "width": 584
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot PCA on 2D\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(8,6))\n",
    "\n",
    "pca_components = range(1, pca.n_components_+1)\n",
    "\n",
    "ax.plot(pca_components, pca.explained_variance_ratio_, lw=2, label='Explained Variance')\n",
    "ax.plot(pca_components, np.cumsum(pca.explained_variance_ratio_), lw=2, label='Cumulative Sum of Explained Variance')\n",
    "\n",
    "ax.set_xlabel('Principal Components')\n",
    "ax.set_ylabel('Explained Variance')\n",
    "ax.set_title('Doc2vec (unigram DBOW + trigram DMM) : explained variance of components')\n",
    "\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In the above graph, the red line represents the cumulative sum of explained variance and the blue line represents the explained variance of each principal component. \n",
    "\n",
    "By looking at the graph above, even though the red line is not perfectly linear, but very close to a straight line. Is this good? No. This means each of the principal components contributes to the variance explanation almost equally and there's no point in reducing the dimensions based on PCA. This can also be seen from the blue line, which is very close to a straight line in the bottom."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "del val_vecs_dbow_dmm_scaled"
   ]
  }
 ],
 "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": "3.7.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}