01_mln-baselines/amazon-electronic/amazon-electronic-descriptive-analysis.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "source": [
    "# Amazon electronic procducts descriptive analysis\n",
    "\n",
    "- Source: Amazon Reviews data (http://jmcauley.ucsd.edu/data/amazon/)\n",
    "\n",
    "The repository has several datasets. For this case study, we are using the Electronics\n",
    "dataset.\n"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "outputs": [],
   "source": [
    "import os\n",
    "import json\n",
    "import gzip\n",
    "from tqdm.notebook import tqdm as tqdm\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "RANDOM = 2021\n",
    "\n",
    "figure_path = '/home/weiss/git/thesis/doc/figures/'"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "outputs": [],
   "source": [
    "DIR = '/home/weiss/rs_data/amazon-electronic-product-recommendation/'\n",
    "PATH_IN = os.path.join(DIR, 'Electronics.json.gz')\n",
    "PATH_META = os.path.join(DIR, 'meta_Electronics.json.gz')"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "outputs": [],
   "source": [
    "### load the meta data\n",
    "\n",
    "def load_data(file, limit_data=False):\n",
    "    data = []\n",
    "    with gzip.open(file) as f:\n",
    "        i = 0\n",
    "        for l in tqdm(f):\n",
    "            data.append(json.loads(l.strip()))\n",
    "            if i >= 100000 and limit_data == True:\n",
    "                break\n",
    "            i +=1\n",
    "    return pd.DataFrame.from_dict(data)"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "outputs": [
    {
     "data": {
      "text/plain": "HBox(children=(HTML(value=''), FloatProgress(value=1.0, bar_style='info', layout=Layout(width='20px'), max=1.0…",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "bb2226d13a5a445095ddb10bafecebd0"
      }
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "   overall  verified   reviewTime      reviewerID        asin  \\\n",
      "0      5.0      True  07 17, 2002  A1N070NS9CJQ2I  0060009810   \n",
      "1      5.0     False   07 6, 2002  A3P0KRKOBQK1KN  0060009810   \n",
      "2      5.0     False   07 3, 2002  A192HO2ICJ75VU  0060009810   \n",
      "3      4.0     False  06 30, 2002  A2T278FKFL3BLT  0060009810   \n",
      "4      5.0     False  06 28, 2002  A2ZUXVTW8RXBXW  0060009810   \n",
      "\n",
      "                       style reviewerName  \\\n",
      "0  {'Format:': ' Hardcover'}   Teri Adams   \n",
      "1  {'Format:': ' Hardcover'}     Willa C.   \n",
      "2  {'Format:': ' Hardcover'}          Kit   \n",
      "3  {'Format:': ' Hardcover'}       Andres   \n",
      "4  {'Format:': ' Hardcover'}         John   \n",
      "\n",
      "                                          reviewText  \\\n",
      "0  This was the first time I read Garcia-Aguilera...   \n",
      "1  As with all of Ms. Garcia-Aguilera's books, I ...   \n",
      "2  I've not read any of Ms Aguilera's works befor...   \n",
      "3  This romance novel is right up there with the ...   \n",
      "4  Carolina Garcia Aguilera has done it again.  S...   \n",
      "\n",
      "                          summary  unixReviewTime vote image  \n",
      "0                   Hit The Spot!      1026864000  NaN   NaN  \n",
      "1  one hot summer is HOT HOT HOT!      1025913600  NaN   NaN  \n",
      "2                  One Hot Summer      1025654400    2   NaN  \n",
      "3               I love this book!      1025395200    3   NaN  \n",
      "4                    One Hot Book      1025222400  NaN   NaN  \n",
      "(20994353, 12)\n"
     ]
    }
   ],
   "source": [
    "ratings = load_data(PATH_IN)\n",
    "print(ratings.head())\n",
    "print(ratings.shape)"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "outputs": [
    {
     "data": {
      "text/plain": "HBox(children=(HTML(value=''), FloatProgress(value=1.0, bar_style='info', layout=Layout(width='20px'), max=1.0…",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "5fd9e83132954816b770d8325ffbc4cf"
      }
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "                                            category tech1  \\\n",
      "0  [Electronics, Camera &amp; Photo, Video Survei...         \n",
      "1                  [Electronics, Camera &amp; Photo]         \n",
      "2  [Electronics, eBook Readers &amp; Accessories,...         \n",
      "3  [Electronics, eBook Readers & Accessories, eBo...         \n",
      "4  [Electronics, eBook Readers & Accessories, eBo...         \n",
      "\n",
      "                                         description fit  \\\n",
      "0  [The following camera brands and models have b...       \n",
      "1  [This second edition of the Handbook of Astron...       \n",
      "2  [A zesty tale. (Publishers Weekly)<br /><br />...       \n",
      "3                                                 []       \n",
      "4  [&#8220;sex.lies.murder.fame. is brillllli&#82...       \n",
      "\n",
      "                                               title  \\\n",
      "0  Genuine Geovision 1 Channel 3rd Party NVR IP S...   \n",
      "1  Books \"Handbook of Astronomical Image Processi...   \n",
      "2                                     One Hot Summer   \n",
      "3  Hurray for Hattie Rabbit: Story and pictures (...   \n",
      "4                     sex.lies.murder.fame.: A Novel   \n",
      "\n",
      "                               also_buy  \\\n",
      "0                                    []   \n",
      "1                          [0999470906]   \n",
      "2              [0425167798, 039914157X]   \n",
      "3  [0060219521, 0060219580, 0060219394]   \n",
      "4                                    []   \n",
      "\n",
      "                                               image tech2  \\\n",
      "0  [https://images-na.ssl-images-amazon.com/image...         \n",
      "1  [https://images-na.ssl-images-amazon.com/image...         \n",
      "2                                                 []         \n",
      "3                                                 []         \n",
      "4                                                 []         \n",
      "\n",
      "                                          brand  \\\n",
      "0                                     GeoVision   \n",
      "1                                  33 Books Co.   \n",
      "2  Visit Amazon's Carolina Garcia Aguilera Page   \n",
      "3           Visit Amazon's Dick Gackenbach Page   \n",
      "4              Visit Amazon's Lolita Files Page   \n",
      "\n",
      "                                             feature  \\\n",
      "0  [Genuine Geovision 1 Channel NVR IP Software, ...   \n",
      "1  [Detailed chapters cover these fundamental top...   \n",
      "2                                                 []   \n",
      "3                                                 []   \n",
      "4                                                 []   \n",
      "\n",
      "                                                rank  \\\n",
      "0  [>#3,092 in Tools &amp; Home Improvement &gt; ...   \n",
      "1  [>#55,933 in Camera &amp; Photo (See Top 100 i...   \n",
      "2                               3,105,177 in Books (   \n",
      "3                               2,024,298 in Books (   \n",
      "4                               3,778,828 in Books (   \n",
      "\n",
      "                              also_view            main_cat similar_item  \\\n",
      "0                                    []  Camera &amp; Photo                \n",
      "1  [0943396670, 1138055360, 0999470906]  Camera &amp; Photo                \n",
      "2                                    []               Books                \n",
      "3  [0060219521, 0060219475, 0060219394]               Books                \n",
      "4                                    []               Books                \n",
      "\n",
      "               date                                              price  \\\n",
      "0  January 28, 2014                                             $65.00   \n",
      "1     June 17, 2003                                                      \n",
      "2                                                               $11.49   \n",
      "3                    .a-section.a-spacing-mini{margin-bottom:6px!im...   \n",
      "4                                                               $13.95   \n",
      "\n",
      "         asin details  \n",
      "0  0011300000     NaN  \n",
      "1  0043396828     NaN  \n",
      "2  0060009810     NaN  \n",
      "3  0060219602     NaN  \n",
      "4  0060786817     NaN  \n",
      "(100001, 18)\n"
     ]
    }
   ],
   "source": [
    "meta = load_data(PATH_META, limit_data=True)\n",
    "print(meta.head())\n",
    "print(meta.shape)"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "outputs": [
    {
     "data": {
      "text/plain": "          user_id     item_id  rating\n0  A1N070NS9CJQ2I  0060009810     5.0\n1  A3P0KRKOBQK1KN  0060009810     5.0\n2  A192HO2ICJ75VU  0060009810     5.0\n3  A2T278FKFL3BLT  0060009810     4.0\n4  A2ZUXVTW8RXBXW  0060009810     5.0",
      "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>user_id</th>\n      <th>item_id</th>\n      <th>rating</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>A1N070NS9CJQ2I</td>\n      <td>0060009810</td>\n      <td>5.0</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>A3P0KRKOBQK1KN</td>\n      <td>0060009810</td>\n      <td>5.0</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>A192HO2ICJ75VU</td>\n      <td>0060009810</td>\n      <td>5.0</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>A2T278FKFL3BLT</td>\n      <td>0060009810</td>\n      <td>4.0</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>A2ZUXVTW8RXBXW</td>\n      <td>0060009810</td>\n      <td>5.0</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "processed_ratings = ratings[['reviewerID', 'asin','overall']]\n",
    "processed_ratings.columns = ['user_id', 'item_id', 'rating']\n",
    "display(processed_ratings.head())"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "# users:  9838676\n",
      "# items:  756489\n",
      "# ratings:  20994353\n"
     ]
    }
   ],
   "source": [
    "print('# users: ', processed_ratings['user_id'].unique().size)\n",
    "print('# items: ', processed_ratings['item_id'].unique().size)\n",
    "print('# ratings: ', processed_ratings['rating'].size)"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "user_id\n",
      "A680RUE1FDO8B     633\n",
      "A3OXHLG6DIBRW8    593\n",
      "ADLVFFE4VBT8      549\n",
      "A1X1CEGHTHMBL1    498\n",
      "A6FIAB28IS79      491\n",
      "                 ... \n",
      "A2JNTVQIKGB5O1      1\n",
      "A2JNTURHIY1LIN      1\n",
      "A2JNTSQZZE069Q      1\n",
      "A2JNTLW8Q6SFKZ      1\n",
      "A2W4MUCX8DW4Q9      1\n",
      "Length: 9838676, dtype: int64\n",
      "<class 'pandas.core.series.Series'>\n",
      "9838676\n"
     ]
    },
    {
     "data": {
      "text/plain": "<Figure size 360x360 with 1 Axes>",
      "image/png": "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\n"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# power law?\n",
    "rating_count_by_user = processed_ratings.groupby('user_id').size().sort_values(ascending=False)\n",
    "print(rating_count_by_user)\n",
    "print(type(rating_count_by_user))\n",
    "size = rating_count_by_user.size\n",
    "print(size)\n",
    "\n",
    "df = pd.DataFrame(dict(sorted_user_id=range(1,size+1), rating_counts=rating_count_by_user.values))\n",
    "\n",
    "sns.set(style='whitegrid', font_scale=1.3, rc={'xtick.labelsize':12, 'ytick.labelsize':12})\n",
    "image = sns.relplot(data=df,\n",
    "                    x=\"sorted_user_id\",\n",
    "                    y=\"rating_counts\",\n",
    "                    kind=\"line\",\n",
    "                    )\n",
    "\n",
    "image.set(xlabel=\"users sorted desc. by # ratings\",\n",
    "          ylabel=\"# ratings\"\n",
    "          )\n",
    "\n",
    "\n",
    "image.savefig(figure_path + '/amazon-long-tail-distribution.png', dpi=300, bbox_inches='tight')"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "False    9833533\n",
      "True        5143\n",
      "Name: user_id, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "# Keep entries where the user has rated more than n items and less than m items\n",
    "\n",
    "n = 50\n",
    "m = 1000\n",
    "\n",
    "counts = processed_ratings['user_id'].value_counts()\n",
    "mask = (counts >= n) & (counts <= m)\n",
    "print(mask.value_counts())\n",
    "processed_ratings_cut = processed_ratings[processed_ratings['user_id'].isin(mask[mask == True].index)]"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "user_id\n",
      "A680RUE1FDO8B     633\n",
      "A3OXHLG6DIBRW8    593\n",
      "ADLVFFE4VBT8      549\n",
      "A1X1CEGHTHMBL1    498\n",
      "A6FIAB28IS79      491\n",
      "                 ... \n",
      "AL3JCDQ3DTGFE      50\n",
      "A2Y3FNMRNCDFUA     50\n",
      "A26OALL5BYR5UH     50\n",
      "A3RKGFB4T4NBCL     50\n",
      "AFZLQ3QNRGOZM      50\n",
      "Length: 5143, dtype: int64\n",
      "<class 'pandas.core.series.Series'>\n",
      "5143\n"
     ]
    },
    {
     "data": {
      "text/plain": "<Figure size 360x360 with 1 Axes>",
      "image/png": "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\n"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# power law?\n",
    "rating_count_by_user = processed_ratings_cut.groupby('user_id').size().sort_values(ascending=False)\n",
    "print(rating_count_by_user)\n",
    "print(type(rating_count_by_user))\n",
    "size = rating_count_by_user.size\n",
    "print(size)\n",
    "\n",
    "df = pd.DataFrame(dict(sorted_user_id=range(1,size+1), rating_counts=rating_count_by_user.values))\n",
    "\n",
    "sns.set(style='whitegrid', font_scale=1.3, rc={'xtick.labelsize':12, 'ytick.labelsize':12})\n",
    "image = sns.relplot(data=df,\n",
    "                    x=\"sorted_user_id\",\n",
    "                    y=\"rating_counts\",\n",
    "                    kind=\"line\",\n",
    "                    )\n",
    "\n",
    "image.set(xlabel=\"users sorted desc. by # ratings\",\n",
    "          ylabel=\"# ratings\"\n",
    "          )\n",
    "\n",
    "\n",
    "image.savefig(figure_path + '/amazon-long-tail-distribution-cut.png', dpi=300, bbox_inches='tight')"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "222         4\n",
      "306         5\n",
      "380         5\n",
      "649         5\n",
      "743         3\n",
      "           ..\n",
      "20993431    5\n",
      "20993438    5\n",
      "20993495    5\n",
      "20993689    4\n",
      "20993740    4\n",
      "Name: rating, Length: 394059, dtype: int64\n",
      "394059\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/weiss/.conda/envs/GNN/lib/python3.7/site-packages/ipykernel_launcher.py:1: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  \"\"\"Entry point for launching an IPython kernel.\n"
     ]
    }
   ],
   "source": [
    "processed_ratings_cut['rating'] = processed_ratings_cut['rating'].values.astype(int)\n",
    "\n",
    "print(processed_ratings_cut['rating'])\n",
    "print(processed_ratings_cut['rating'].size)"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/weiss/.conda/envs/GNN/lib/python3.7/site-packages/seaborn/categorical.py:3704: UserWarning: The `factorplot` function has been renamed to `catplot`. The original name will be removed in a future release. Please update your code. Note that the default `kind` in `factorplot` (`'point'`) has changed `'strip'` in `catplot`.\n",
      "  warnings.warn(msg)\n",
      "/home/weiss/.conda/envs/GNN/lib/python3.7/site-packages/seaborn/_decorators.py:43: 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",
      "  FutureWarning\n"
     ]
    },
    {
     "data": {
      "text/plain": "<Figure size 360x360 with 1 Axes>",
      "image/png": "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\n"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# rating distribution after cleanup\n",
    "sns.set(style='whitegrid', font_scale=1.3)\n",
    "g = sns.factorplot(\"rating\", data=processed_ratings_cut, kind='count')\n",
    "g.set_ylabels(\"# ratings\")\n",
    "g.savefig(figure_path + '/amazon-rating-distribution.png', dpi=300, bbox_inches='tight')"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "outputs": [
    {
     "data": {
      "text/plain": "4.340865200388774"
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "processed_ratings_cut['rating'].mean()"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "outputs": [
    {
     "data": {
      "text/plain": "5.0"
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "processed_ratings_cut['rating'].median()"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "outputs": [],
   "source": [
    "# save processed ratings as pickle\n",
    "processed_ratings_cut.to_pickle(DIR+'amazon-electronic-ratings.pkl')"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "outputs": [],
   "source": [],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  }
 ],
 "metadata": {
  "kernelspec": {
   "name": "pycharm-f57757a1",
   "language": "python",
   "display_name": "PyCharm (rs-via-gnn)"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.9"
  },
  "papermill": {
   "duration": 38.668056,
   "end_time": "2021-01-16T05:36:39.152998",
   "environment_variables": {},
   "exception": null,
   "input_path": "__notebook__.ipynb",
   "output_path": "__notebook__.ipynb",
   "parameters": {},
   "start_time": "2021-01-16T05:36:00.484942",
   "version": "2.1.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}