create_data.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from random import sample, seed\n",
    "\n",
    "from matplotlib import pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Create Interactions File\n",
    "1. Initially selected download interactions.\n",
    "2. Identified the top 8000 most installed applications.\n",
    "3. Retained interactions related to these top applications.\n",
    "4. Filtered for users with more than 32 interactions.\n",
    "5. Randomly sampled 10,000 users from the filtered group for the final dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('raw_myket_input.csv', sep='\\t', index_col=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(53632867, 4)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>app_name</th>\n",
       "      <th>user_id</th>\n",
       "      <th>datetime</th>\n",
       "      <th>is_update</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>com.cocoplay.erpetvet</td>\n",
       "      <td>392863962</td>\n",
       "      <td>2020-06-17 23:55:17.460</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                app_name    user_id                 datetime  is_update\n",
       "0  com.cocoplay.erpetvet  392863962  2020-06-17 23:55:17.460          0"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_install = df[df['is_update'] == 0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(36615586, 4)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_install.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_igb_user =df_install.groupby('user_id').count()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_igb_app =df_install.groupby('app_name').count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1295520, 3)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_igb_user.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "top_apps = df_igb_app.sort_values('user_id', ascending=False).head(8000).index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_if = df_install[df_install['app_name'].isin(top_apps)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5569723692004009"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_if.shape[0] / df.shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_if_gb_user = df_if.groupby('user_id').count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "top_users = df_if_gb_user[df_if_gb_user['app_name'] > 32].index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_iff = df_if[df_if['user_id'].isin(top_users)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "280391"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(top_users)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.36765345772024455"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_iff.shape[0] / df.shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "seed(42)\n",
    "sample_users = sample(list(top_users), 10000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "10000"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(set(sample_users))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_sample = df_iff[df_iff['user_id'].isin(sample_users)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(694121, 4)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_sample.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(7988, 3)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_sample.groupby('app_name').count().shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create Timestamp based on datetime"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/xv/yt1388s53fg1889wfzl2l37r0000gr/T/ipykernel_38240/1389366755.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",
      "  df_sample['datetime'] = pd.to_datetime(df_sample['datetime'])\n"
     ]
    }
   ],
   "source": [
    "df_sample['datetime'] = pd.to_datetime(df_sample['datetime'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "min_time = df_sample['datetime'].min()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/xv/yt1388s53fg1889wfzl2l37r0000gr/T/ipykernel_38240/4151269595.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",
      "  df_sample['timestamp'] = df_sample['datetime'].apply(lambda x: (x - min_time).total_seconds())\n"
     ]
    }
   ],
   "source": [
    "df_sample['timestamp'] = df_sample['datetime'].apply(lambda x: (x - min_time).total_seconds())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot histogram of timestamps\n",
    "plt.hist(df_sample['timestamp'], bins=100)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/xv/yt1388s53fg1889wfzl2l37r0000gr/T/ipykernel_38240/3304871164.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",
      "  df_sample['is_update2'] = 0\n",
      "/var/folders/xv/yt1388s53fg1889wfzl2l37r0000gr/T/ipykernel_38240/3304871164.py:2: 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",
      "  df_sample['is_update3'] = 0\n"
     ]
    }
   ],
   "source": [
    "df_sample['is_update2'] = 0\n",
    "df_sample['is_update3'] = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_sample[['user_id', 'app_name', 'timestamp', 'is_update', 'is_update2', 'is_update3']].to_csv('myket.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "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>app_name</th>\n",
       "      <th>user_id</th>\n",
       "      <th>datetime</th>\n",
       "      <th>is_update</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>is_update2</th>\n",
       "      <th>is_update3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>97</th>\n",
       "      <td>com.tencent.ig</td>\n",
       "      <td>-1185417981</td>\n",
       "      <td>2020-06-17 23:56:04.440</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>546</th>\n",
       "      <td>com.tencent.iglite</td>\n",
       "      <td>-562407438</td>\n",
       "      <td>2020-06-17 23:59:44.787</td>\n",
       "      <td>0</td>\n",
       "      <td>220.347</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               app_name     user_id                datetime  is_update  \\\n",
       "97       com.tencent.ig -1185417981 2020-06-17 23:56:04.440          0   \n",
       "546  com.tencent.iglite  -562407438 2020-06-17 23:59:44.787          0   \n",
       "\n",
       "     timestamp  is_update2  is_update3  \n",
       "97       0.000           0           0  \n",
       "546    220.347           0           0  "
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_sample.head(2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Create Interactions File with integer identifiers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/xv/yt1388s53fg1889wfzl2l37r0000gr/T/ipykernel_38240/4079327224.py:4: 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",
      "  df_sample['app_name_id'] = df_sample['app_name'].map(app_names_mapping)\n"
     ]
    }
   ],
   "source": [
    "app_names = df_sample['app_name'].unique()\n",
    "app_names.sort()\n",
    "app_names_mapping = {app_name: i for i, app_name in enumerate(app_names)}\n",
    "df_sample['app_name_id'] = df_sample['app_name'].map(app_names_mapping)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/xv/yt1388s53fg1889wfzl2l37r0000gr/T/ipykernel_38240/3047062015.py:4: 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",
      "  df_sample['user_id_map'] = df_sample['user_id'].map(user_ids_mapping)\n"
     ]
    }
   ],
   "source": [
    "user_ids = df_sample['user_id'].unique()\n",
    "user_ids.sort()\n",
    "user_ids_mapping = {user_id: i for i, user_id in enumerate(user_ids)}\n",
    "df_sample['user_id_map'] = df_sample['user_id'].map(user_ids_mapping)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_sample[['user_id_map', 'app_name_id', 'timestamp', 'is_update', 'is_update2', 'is_update3']].to_csv('data_int_index/myket.csv', index=False)\n",
    "pd.DataFrame({'app_name': app_names, 'app_name_id': list(app_names_mapping.values())}).to_csv('data_int_index/app_name_mapping.csv', index=False)\n",
    "pd.DataFrame({'user_id': user_ids, 'user_id_map': list(user_ids_mapping.values())}).to_csv('data_int_index/user_id_mapping.csv', index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Create Application Info File"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "app_info = pd.read_csv('app_info.csv', sep='\\t', header=None, names=[\"app_name\", \"installs\", \"rating\", \"rating_count\", \"category_fa\"], index_col=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "categories = pd.read_csv('categories.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "app_info = app_info.merge(categories, on='category_fa', how='left')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "app_info_sample = app_info[app_info['app_name'].isin(df_sample['app_name'].unique())]\n",
    "app_info_sample.to_csv('app_info_sample.csv', index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Generate Numpy Version of Application Info"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "cat_dummies = pd.get_dummies(app_info_sample['category_en'])\n",
    "app_info_sample = pd.concat([app_info_sample, cat_dummies], axis=1)\n",
    "app_info_sample['app_name_id'] = app_info_sample['app_name'].map(app_names_mapping)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "all_apps = pd.DataFrame({'app_name_id': np.arange(len(app_names))})\n",
    "app_info_sample = pd.merge(all_apps, app_info_sample, on='app_name_id', how='left')\n",
    "app_info_sample = app_info_sample.sort_values(by='app_name_id')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "app_info_tab = app_info_sample[['installs', 'rating', 'rating_count'] + list(cat_dummies.columns)]\n",
    "app_info_np = app_info_tab.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "np.save('data_int_index/app_info_sample.npy', app_info_np)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Generate Interaction file with integer identifiers and features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_sample_np = df_sample[['user_id_map', 'app_name_id', 'timestamp', 'is_update']].values\n",
    "df_sample_featuers = app_info_np[df_sample_np[:, 1].astype(int)]\n",
    "df_sample_featuers = np.nan_to_num(df_sample_featuers, nan=0.0)\n",
    "df_sample_np = np.concatenate([df_sample_np, df_sample_featuers], axis=1)\n",
    "np.savetxt('data_int_index/myket_with_features.csv', df_sample_np, delimiter=',', header='user_id_map,app_name_id,timestamp,state_label,comma_separated_list_of_features', fmt=\"%10.5f\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.9.13 ('base')",
   "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": "3.9.13"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "34038ecf986242e43f92b2b60c5b9ff39d8e56de39e9b83b743c24dd3412e538"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}