Anime Recommender.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Anime Recommender\n",
    "\n",
    "*Recommending similar anime using KNN and K-Means*\n",
    "\n",
    "\n",
    "CSE445 Course Project and my first Machine Learning Project."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "#importing libraries\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import math\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from sklearn.neighbors import NearestNeighbors\n",
    "from sklearn.cluster import KMeans\n",
    "import seaborn as sns\n",
    "from sklearn import metrics\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "from IPython.display import clear_output\n",
    "import time\n",
    "from sklearn.metrics import silhouette_score\n",
    "from sklearn.metrics import davies_bouldin_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "#source dataset path\n",
    "path1 = r'D:\\NSU\\11th Semester\\CSE445 - AZK\\Project\\Anime.csv'\n",
    "path2 = r'D:\\NSU\\11th Semester\\CSE445 - AZK\\Project\\profiles.csv'"
   ]
  },
  {
   "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>animeID</th>\n",
       "      <th>name</th>\n",
       "      <th>title_english</th>\n",
       "      <th>type</th>\n",
       "      <th>source</th>\n",
       "      <th>producers</th>\n",
       "      <th>genre</th>\n",
       "      <th>studio</th>\n",
       "      <th>episodes</th>\n",
       "      <th>score</th>\n",
       "      <th>scored_by</th>\n",
       "      <th>popularity</th>\n",
       "      <th>members</th>\n",
       "      <th>favorites</th>\n",
       "      <th>synopsis</th>\n",
       "      <th>premiered</th>\n",
       "      <th>related</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Cowboy Bebop</td>\n",
       "      <td>Cowboy Bebop</td>\n",
       "      <td>TV</td>\n",
       "      <td>Original</td>\n",
       "      <td>['Bandai Visual']</td>\n",
       "      <td>['Action', 'Adventure', 'Comedy', 'Drama', 'Sc...</td>\n",
       "      <td>['Sunrise']</td>\n",
       "      <td>26.0</td>\n",
       "      <td>8.81</td>\n",
       "      <td>405664.0</td>\n",
       "      <td>39.0</td>\n",
       "      <td>795733.0</td>\n",
       "      <td>43460.0</td>\n",
       "      <td>In the year 2071, humanity has colonized sever...</td>\n",
       "      <td>Spring 1998</td>\n",
       "      <td>{'Adaptation': [{'mal_id': 173, 'type': 'manga...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>Cowboy Bebop: Tengoku no Tobira</td>\n",
       "      <td>Cowboy Bebop: The Movie</td>\n",
       "      <td>Movie</td>\n",
       "      <td>Original</td>\n",
       "      <td>['Sunrise', 'Bandai Visual']</td>\n",
       "      <td>['Action', 'Drama', 'Mystery', 'Sci-Fi', 'Space']</td>\n",
       "      <td>['Bones']</td>\n",
       "      <td>1.0</td>\n",
       "      <td>8.41</td>\n",
       "      <td>120243.0</td>\n",
       "      <td>449.0</td>\n",
       "      <td>197791.0</td>\n",
       "      <td>776.0</td>\n",
       "      <td>Another day, another bounty—such is the life o...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'Parent story': [{'mal_id': 1, 'type': 'anime...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>Trigun</td>\n",
       "      <td>Trigun</td>\n",
       "      <td>TV</td>\n",
       "      <td>Manga</td>\n",
       "      <td>['Victor Entertainment']</td>\n",
       "      <td>['Action', 'Sci-Fi', 'Adventure', 'Comedy', 'D...</td>\n",
       "      <td>['Madhouse']</td>\n",
       "      <td>26.0</td>\n",
       "      <td>8.30</td>\n",
       "      <td>212537.0</td>\n",
       "      <td>146.0</td>\n",
       "      <td>408548.0</td>\n",
       "      <td>10432.0</td>\n",
       "      <td>Vash the Stampede is the man with a $$60,000,0...</td>\n",
       "      <td>Spring 1998</td>\n",
       "      <td>{'Adaptation': [{'mal_id': 703, 'type': 'manga...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>7</td>\n",
       "      <td>Witch Hunter Robin</td>\n",
       "      <td>Witch Hunter Robin</td>\n",
       "      <td>TV</td>\n",
       "      <td>Original</td>\n",
       "      <td>['Bandai Visual']</td>\n",
       "      <td>['Action', 'Magic', 'Police', 'Supernatural', ...</td>\n",
       "      <td>['Sunrise']</td>\n",
       "      <td>26.0</td>\n",
       "      <td>7.33</td>\n",
       "      <td>32837.0</td>\n",
       "      <td>1171.0</td>\n",
       "      <td>79397.0</td>\n",
       "      <td>537.0</td>\n",
       "      <td>Witches are individuals with special powers li...</td>\n",
       "      <td>Summer 2002</td>\n",
       "      <td>{}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>8</td>\n",
       "      <td>Bouken Ou Beet</td>\n",
       "      <td>Beet the Vandel Buster</td>\n",
       "      <td>TV</td>\n",
       "      <td>Manga</td>\n",
       "      <td>['TV Tokyo', 'Dentsu']</td>\n",
       "      <td>['Adventure', 'Fantasy', 'Shounen', 'Supernatu...</td>\n",
       "      <td>['Toei Animation']</td>\n",
       "      <td>52.0</td>\n",
       "      <td>7.03</td>\n",
       "      <td>4894.0</td>\n",
       "      <td>3704.0</td>\n",
       "      <td>11708.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>It is the dark century and the people are suff...</td>\n",
       "      <td>Fall 2004</td>\n",
       "      <td>{'Adaptation': [{'mal_id': 1348, 'type': 'mang...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   animeID                             name            title_english   type  \\\n",
       "0        1                     Cowboy Bebop             Cowboy Bebop     TV   \n",
       "1        5  Cowboy Bebop: Tengoku no Tobira  Cowboy Bebop: The Movie  Movie   \n",
       "2        6                           Trigun                   Trigun     TV   \n",
       "3        7               Witch Hunter Robin       Witch Hunter Robin     TV   \n",
       "4        8                   Bouken Ou Beet   Beet the Vandel Buster     TV   \n",
       "\n",
       "     source                     producers  \\\n",
       "0  Original             ['Bandai Visual']   \n",
       "1  Original  ['Sunrise', 'Bandai Visual']   \n",
       "2     Manga      ['Victor Entertainment']   \n",
       "3  Original             ['Bandai Visual']   \n",
       "4     Manga        ['TV Tokyo', 'Dentsu']   \n",
       "\n",
       "                                               genre              studio  \\\n",
       "0  ['Action', 'Adventure', 'Comedy', 'Drama', 'Sc...         ['Sunrise']   \n",
       "1  ['Action', 'Drama', 'Mystery', 'Sci-Fi', 'Space']           ['Bones']   \n",
       "2  ['Action', 'Sci-Fi', 'Adventure', 'Comedy', 'D...        ['Madhouse']   \n",
       "3  ['Action', 'Magic', 'Police', 'Supernatural', ...         ['Sunrise']   \n",
       "4  ['Adventure', 'Fantasy', 'Shounen', 'Supernatu...  ['Toei Animation']   \n",
       "\n",
       "   episodes  score  scored_by  popularity   members  favorites  \\\n",
       "0      26.0   8.81   405664.0        39.0  795733.0    43460.0   \n",
       "1       1.0   8.41   120243.0       449.0  197791.0      776.0   \n",
       "2      26.0   8.30   212537.0       146.0  408548.0    10432.0   \n",
       "3      26.0   7.33    32837.0      1171.0   79397.0      537.0   \n",
       "4      52.0   7.03     4894.0      3704.0   11708.0       14.0   \n",
       "\n",
       "                                            synopsis    premiered  \\\n",
       "0  In the year 2071, humanity has colonized sever...  Spring 1998   \n",
       "1  Another day, another bounty—such is the life o...          NaN   \n",
       "2  Vash the Stampede is the man with a $$60,000,0...  Spring 1998   \n",
       "3  Witches are individuals with special powers li...  Summer 2002   \n",
       "4  It is the dark century and the people are suff...    Fall 2004   \n",
       "\n",
       "                                             related  \n",
       "0  {'Adaptation': [{'mal_id': 173, 'type': 'manga...  \n",
       "1  {'Parent story': [{'mal_id': 1, 'type': 'anime...  \n",
       "2  {'Adaptation': [{'mal_id': 703, 'type': 'manga...  \n",
       "3                                                 {}  \n",
       "4  {'Adaptation': [{'mal_id': 1348, 'type': 'mang...  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#reading csv file\n",
    "anime_data = pd.read_csv(path1)\n",
    "anime_data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Preparing Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['animeID', 'name', 'title_english', 'type', 'source', 'producers',\n",
       "       'genre', 'studio', 'episodes', 'score', 'scored_by', 'popularity',\n",
       "       'members', 'favorites', 'synopsis', 'premiered', 'related'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "anime_data.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "#removing unnecessary Columns\n",
    "del anime_data['producers']\n",
    "del anime_data['studio']\n",
    "del anime_data['popularity']\n",
    "del anime_data['members']\n",
    "del anime_data['favorites']\n",
    "del anime_data['synopsis']\n",
    "del anime_data['premiered']\n",
    "del anime_data['related']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>animeID</th>\n",
       "      <th>name</th>\n",
       "      <th>title_english</th>\n",
       "      <th>type</th>\n",
       "      <th>source</th>\n",
       "      <th>genre</th>\n",
       "      <th>episodes</th>\n",
       "      <th>score</th>\n",
       "      <th>scored_by</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Cowboy Bebop</td>\n",
       "      <td>Cowboy Bebop</td>\n",
       "      <td>TV</td>\n",
       "      <td>Original</td>\n",
       "      <td>['Action', 'Adventure', 'Comedy', 'Drama', 'Sc...</td>\n",
       "      <td>26.0</td>\n",
       "      <td>8.81</td>\n",
       "      <td>405664.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>Cowboy Bebop: Tengoku no Tobira</td>\n",
       "      <td>Cowboy Bebop: The Movie</td>\n",
       "      <td>Movie</td>\n",
       "      <td>Original</td>\n",
       "      <td>['Action', 'Drama', 'Mystery', 'Sci-Fi', 'Space']</td>\n",
       "      <td>1.0</td>\n",
       "      <td>8.41</td>\n",
       "      <td>120243.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>Trigun</td>\n",
       "      <td>Trigun</td>\n",
       "      <td>TV</td>\n",
       "      <td>Manga</td>\n",
       "      <td>['Action', 'Sci-Fi', 'Adventure', 'Comedy', 'D...</td>\n",
       "      <td>26.0</td>\n",
       "      <td>8.30</td>\n",
       "      <td>212537.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>7</td>\n",
       "      <td>Witch Hunter Robin</td>\n",
       "      <td>Witch Hunter Robin</td>\n",
       "      <td>TV</td>\n",
       "      <td>Original</td>\n",
       "      <td>['Action', 'Magic', 'Police', 'Supernatural', ...</td>\n",
       "      <td>26.0</td>\n",
       "      <td>7.33</td>\n",
       "      <td>32837.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>8</td>\n",
       "      <td>Bouken Ou Beet</td>\n",
       "      <td>Beet the Vandel Buster</td>\n",
       "      <td>TV</td>\n",
       "      <td>Manga</td>\n",
       "      <td>['Adventure', 'Fantasy', 'Shounen', 'Supernatu...</td>\n",
       "      <td>52.0</td>\n",
       "      <td>7.03</td>\n",
       "      <td>4894.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   animeID                             name            title_english   type  \\\n",
       "0        1                     Cowboy Bebop             Cowboy Bebop     TV   \n",
       "1        5  Cowboy Bebop: Tengoku no Tobira  Cowboy Bebop: The Movie  Movie   \n",
       "2        6                           Trigun                   Trigun     TV   \n",
       "3        7               Witch Hunter Robin       Witch Hunter Robin     TV   \n",
       "4        8                   Bouken Ou Beet   Beet the Vandel Buster     TV   \n",
       "\n",
       "     source                                              genre  episodes  \\\n",
       "0  Original  ['Action', 'Adventure', 'Comedy', 'Drama', 'Sc...      26.0   \n",
       "1  Original  ['Action', 'Drama', 'Mystery', 'Sci-Fi', 'Space']       1.0   \n",
       "2     Manga  ['Action', 'Sci-Fi', 'Adventure', 'Comedy', 'D...      26.0   \n",
       "3  Original  ['Action', 'Magic', 'Police', 'Supernatural', ...      26.0   \n",
       "4     Manga  ['Adventure', 'Fantasy', 'Shounen', 'Supernatu...      52.0   \n",
       "\n",
       "   score  scored_by  \n",
       "0   8.81   405664.0  \n",
       "1   8.41   120243.0  \n",
       "2   8.30   212537.0  \n",
       "3   7.33    32837.0  \n",
       "4   7.03     4894.0  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "anime_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "#removing completely blank cell\n",
    "anime_data = anime_data[anime_data.name != '404']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>animeID</th>\n",
       "      <th>name</th>\n",
       "      <th>title_english</th>\n",
       "      <th>type</th>\n",
       "      <th>source</th>\n",
       "      <th>genre</th>\n",
       "      <th>episodes</th>\n",
       "      <th>score</th>\n",
       "      <th>scored_by</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Cowboy Bebop</td>\n",
       "      <td>Cowboy Bebop</td>\n",
       "      <td>TV</td>\n",
       "      <td>Original</td>\n",
       "      <td>Action, Adventure, Comedy, Drama, Sci-Fi, Space</td>\n",
       "      <td>26.0</td>\n",
       "      <td>8.81</td>\n",
       "      <td>405664.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>Cowboy Bebop: Tengoku no Tobira</td>\n",
       "      <td>Cowboy Bebop: The Movie</td>\n",
       "      <td>Movie</td>\n",
       "      <td>Original</td>\n",
       "      <td>Action, Drama, Mystery, Sci-Fi, Space</td>\n",
       "      <td>1.0</td>\n",
       "      <td>8.41</td>\n",
       "      <td>120243.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>Trigun</td>\n",
       "      <td>Trigun</td>\n",
       "      <td>TV</td>\n",
       "      <td>Manga</td>\n",
       "      <td>Action, Sci-Fi, Adventure, Comedy, Drama, Shounen</td>\n",
       "      <td>26.0</td>\n",
       "      <td>8.30</td>\n",
       "      <td>212537.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>7</td>\n",
       "      <td>Witch Hunter Robin</td>\n",
       "      <td>Witch Hunter Robin</td>\n",
       "      <td>TV</td>\n",
       "      <td>Original</td>\n",
       "      <td>Action, Magic, Police, Supernatural, Drama, My...</td>\n",
       "      <td>26.0</td>\n",
       "      <td>7.33</td>\n",
       "      <td>32837.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>8</td>\n",
       "      <td>Bouken Ou Beet</td>\n",
       "      <td>Beet the Vandel Buster</td>\n",
       "      <td>TV</td>\n",
       "      <td>Manga</td>\n",
       "      <td>Adventure, Fantasy, Shounen, Supernatural</td>\n",
       "      <td>52.0</td>\n",
       "      <td>7.03</td>\n",
       "      <td>4894.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   animeID                             name            title_english   type  \\\n",
       "0        1                     Cowboy Bebop             Cowboy Bebop     TV   \n",
       "1        5  Cowboy Bebop: Tengoku no Tobira  Cowboy Bebop: The Movie  Movie   \n",
       "2        6                           Trigun                   Trigun     TV   \n",
       "3        7               Witch Hunter Robin       Witch Hunter Robin     TV   \n",
       "4        8                   Bouken Ou Beet   Beet the Vandel Buster     TV   \n",
       "\n",
       "     source                                              genre  episodes  \\\n",
       "0  Original    Action, Adventure, Comedy, Drama, Sci-Fi, Space      26.0   \n",
       "1  Original              Action, Drama, Mystery, Sci-Fi, Space       1.0   \n",
       "2     Manga  Action, Sci-Fi, Adventure, Comedy, Drama, Shounen      26.0   \n",
       "3  Original  Action, Magic, Police, Supernatural, Drama, My...      26.0   \n",
       "4     Manga          Adventure, Fantasy, Shounen, Supernatural      52.0   \n",
       "\n",
       "   score  scored_by  \n",
       "0   8.81   405664.0  \n",
       "1   8.41   120243.0  \n",
       "2   8.30   212537.0  \n",
       "3   7.33    32837.0  \n",
       "4   7.03     4894.0  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#removing [,] and ' from genre\n",
    "temp1 = anime_data['genre']\n",
    "\n",
    "temp1 = temp1.str.replace('[','',regex=True)\n",
    "temp1 = temp1.str.replace(']','',regex=True)\n",
    "temp1 = temp1.str.replace(\"'\",'',regex=True)\n",
    "\n",
    "#updating genre\n",
    "anime_data['genre'] = temp1\n",
    "anime_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>animeID</th>\n",
       "      <th>name</th>\n",
       "      <th>title_english</th>\n",
       "      <th>type</th>\n",
       "      <th>source</th>\n",
       "      <th>genre</th>\n",
       "      <th>episodes</th>\n",
       "      <th>score</th>\n",
       "      <th>scored_by</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>8859</td>\n",
       "      <td>24023</td>\n",
       "      <td>Project758</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>Drama</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9611</td>\n",
       "      <td>28613</td>\n",
       "      <td>Ginga Jinpuu Jinraiger</td>\n",
       "      <td>Galaxy Divine Wind Jinraiger</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>Original</td>\n",
       "      <td>Action, Adventure, Mecha</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>10339</td>\n",
       "      <td>30448</td>\n",
       "      <td>Mirai Arise</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>Original</td>\n",
       "      <td>Sci-Fi</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>10344</td>\n",
       "      <td>30455</td>\n",
       "      <td>Kantai Collection: KanColle Zoku-hen</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>Card game</td>\n",
       "      <td>Action, Military, Sci-Fi, Slice of Life, School</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11167</td>\n",
       "      <td>32455</td>\n",
       "      <td>Gekidol</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>Music</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       animeID                                  name  \\\n",
       "8859     24023                            Project758   \n",
       "9611     28613                Ginga Jinpuu Jinraiger   \n",
       "10339    30448                           Mirai Arise   \n",
       "10344    30455  Kantai Collection: KanColle Zoku-hen   \n",
       "11167    32455                               Gekidol   \n",
       "\n",
       "                      title_english     type     source  \\\n",
       "8859                            NaN  Unknown    Unknown   \n",
       "9611   Galaxy Divine Wind Jinraiger  Unknown   Original   \n",
       "10339                           NaN  Unknown   Original   \n",
       "10344                           NaN  Unknown  Card game   \n",
       "11167                           NaN  Unknown    Unknown   \n",
       "\n",
       "                                                 genre  episodes  score  \\\n",
       "8859                                             Drama       NaN    NaN   \n",
       "9611                          Action, Adventure, Mecha       NaN    NaN   \n",
       "10339                                           Sci-Fi       NaN    NaN   \n",
       "10344  Action, Military, Sci-Fi, Slice of Life, School       NaN    NaN   \n",
       "11167                                            Music       NaN    NaN   \n",
       "\n",
       "       scored_by  \n",
       "8859         0.0  \n",
       "9611         0.0  \n",
       "10339        0.0  \n",
       "10344        0.0  \n",
       "11167        1.0  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#checking if there is any unknown cell in type\n",
    "anime_data[anime_data['type']=='Unknown'].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "#removed unknown type animesource\n",
    "anime_data = anime_data[anime_data.type != 'Unknown']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "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>animeID</th>\n",
       "      <th>name</th>\n",
       "      <th>title_english</th>\n",
       "      <th>type</th>\n",
       "      <th>source</th>\n",
       "      <th>genre</th>\n",
       "      <th>episodes</th>\n",
       "      <th>score</th>\n",
       "      <th>scored_by</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>60</td>\n",
       "      <td>88</td>\n",
       "      <td>Mobile Suit Gundam F91</td>\n",
       "      <td>Mobile Suit Gundam F91</td>\n",
       "      <td>Movie</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>Drama, Mecha, Military, Sci-Fi, Space</td>\n",
       "      <td>1.0</td>\n",
       "      <td>6.78</td>\n",
       "      <td>9833.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>78</td>\n",
       "      <td>108</td>\n",
       "      <td>Ou Dorobou Jing in Seventh Heaven</td>\n",
       "      <td>Jing: King of Bandits - Seventh Heaven</td>\n",
       "      <td>OVA</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>Adventure, Fantasy, Psychological, Comedy, Sho...</td>\n",
       "      <td>3.0</td>\n",
       "      <td>7.37</td>\n",
       "      <td>4665.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>140</td>\n",
       "      <td>176</td>\n",
       "      <td>Triangle Heart: Sweet Songs Forever</td>\n",
       "      <td>NaN</td>\n",
       "      <td>OVA</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>Adventure, Romance</td>\n",
       "      <td>4.0</td>\n",
       "      <td>6.24</td>\n",
       "      <td>2182.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>153</td>\n",
       "      <td>192</td>\n",
       "      <td>Love Hina Haru Special: Kimi Sakura Chiru Naka...</td>\n",
       "      <td>Love Hina Spring Movie</td>\n",
       "      <td>Special</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>Adventure, Comedy, Romance</td>\n",
       "      <td>1.0</td>\n",
       "      <td>7.27</td>\n",
       "      <td>27974.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>155</td>\n",
       "      <td>194</td>\n",
       "      <td>Macross Zero</td>\n",
       "      <td>Macross Zero</td>\n",
       "      <td>OVA</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>Adventure, Mecha, Military, Sci-Fi, Shounen</td>\n",
       "      <td>5.0</td>\n",
       "      <td>7.62</td>\n",
       "      <td>15264.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     animeID                                               name  \\\n",
       "60        88                             Mobile Suit Gundam F91   \n",
       "78       108                  Ou Dorobou Jing in Seventh Heaven   \n",
       "140      176                Triangle Heart: Sweet Songs Forever   \n",
       "153      192  Love Hina Haru Special: Kimi Sakura Chiru Naka...   \n",
       "155      194                                       Macross Zero   \n",
       "\n",
       "                              title_english     type   source  \\\n",
       "60                   Mobile Suit Gundam F91    Movie  Unknown   \n",
       "78   Jing: King of Bandits - Seventh Heaven      OVA  Unknown   \n",
       "140                                     NaN      OVA  Unknown   \n",
       "153                  Love Hina Spring Movie  Special  Unknown   \n",
       "155                            Macross Zero      OVA  Unknown   \n",
       "\n",
       "                                                 genre  episodes  score  \\\n",
       "60               Drama, Mecha, Military, Sci-Fi, Space       1.0   6.78   \n",
       "78   Adventure, Fantasy, Psychological, Comedy, Sho...       3.0   7.37   \n",
       "140                                 Adventure, Romance       4.0   6.24   \n",
       "153                         Adventure, Comedy, Romance       1.0   7.27   \n",
       "155        Adventure, Mecha, Military, Sci-Fi, Shounen       5.0   7.62   \n",
       "\n",
       "     scored_by  \n",
       "60      9833.0  \n",
       "78      4665.0  \n",
       "140     2182.0  \n",
       "153    27974.0  \n",
       "155    15264.0  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#checking if there is any unknown cell in source\n",
    "anime_data[anime_data['source']=='Unknown'].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "#assuming unknow sources are original\n",
    "anime_data.loc[anime_data['source']=='Unknown','source'] = 'Original'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "#clearing cells with 0 scored by\n",
    "anime_data.loc[anime_data['scored_by']==0,'scored_by'] = None\n",
    "#anime_data[anime_data['scored_by']==0].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['TV', 'Movie', 'OVA', 'Special', 'ONA', 'Music'], dtype=object)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#checking all types for anime  \n",
    "anime_data.type.unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "24.0"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#finding median of tv series episodes\n",
    "tv = anime_data[anime_data['type'] == 'TV']\n",
    "tv_median = tv['episodes'].median(skipna=True)\n",
    "tv_median"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "D:\\Programs\\Anaconda3\\lib\\site-packages\\pandas\\core\\generic.py:6287: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  self._update_inplace(new_data)\n"
     ]
    }
   ],
   "source": [
    "#filling tv episodes null value with median\n",
    "tv['episodes'].fillna(tv_median, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "#finding average tv rating and average scored by\n",
    "tv_rating_avg = tv['score'].mean(skipna=True)\n",
    "tv_scoredBy_avg = math.floor(tv['scored_by'].mean(skipna=True))\n",
    "\n",
    "#filling empty tv rating and scored by with average\n",
    "tv['score'].fillna(tv_rating_avg, inplace = True)\n",
    "tv['scored_by'].fillna(tv_scoredBy_avg, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "#merging these values in main anime_data\n",
    "anime_data.update(tv)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "#not needed anymore\n",
    "#replacing any unknown cell in episode with 1\n",
    "#anime_data['episodes'].fillna(1, inplace = True)\n",
    "#anime_data[anime_data['episodes']==1].head()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "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>animeID</th>\n",
       "      <th>name</th>\n",
       "      <th>title_english</th>\n",
       "      <th>type</th>\n",
       "      <th>source</th>\n",
       "      <th>genre</th>\n",
       "      <th>episodes</th>\n",
       "      <th>score</th>\n",
       "      <th>scored_by</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Cowboy Bebop</td>\n",
       "      <td>Cowboy Bebop</td>\n",
       "      <td>TV</td>\n",
       "      <td>Original</td>\n",
       "      <td>Action, Adventure, Comedy, Drama, Sci-Fi, Space</td>\n",
       "      <td>26.0</td>\n",
       "      <td>8.81</td>\n",
       "      <td>405664.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>Trigun</td>\n",
       "      <td>Trigun</td>\n",
       "      <td>TV</td>\n",
       "      <td>Manga</td>\n",
       "      <td>Action, Sci-Fi, Adventure, Comedy, Drama, Shounen</td>\n",
       "      <td>26.0</td>\n",
       "      <td>8.30</td>\n",
       "      <td>212537.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>7</td>\n",
       "      <td>Witch Hunter Robin</td>\n",
       "      <td>Witch Hunter Robin</td>\n",
       "      <td>TV</td>\n",
       "      <td>Original</td>\n",
       "      <td>Action, Magic, Police, Supernatural, Drama, My...</td>\n",
       "      <td>26.0</td>\n",
       "      <td>7.33</td>\n",
       "      <td>32837.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>8</td>\n",
       "      <td>Bouken Ou Beet</td>\n",
       "      <td>Beet the Vandel Buster</td>\n",
       "      <td>TV</td>\n",
       "      <td>Manga</td>\n",
       "      <td>Adventure, Fantasy, Shounen, Supernatural</td>\n",
       "      <td>52.0</td>\n",
       "      <td>7.03</td>\n",
       "      <td>4894.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>16</td>\n",
       "      <td>Hachimitsu to Clover</td>\n",
       "      <td>Honey and Clover</td>\n",
       "      <td>TV</td>\n",
       "      <td>Manga</td>\n",
       "      <td>Comedy, Drama, Josei, Romance, Slice of Life</td>\n",
       "      <td>24.0</td>\n",
       "      <td>8.12</td>\n",
       "      <td>57065.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   animeID                  name           title_english type    source  \\\n",
       "0        1          Cowboy Bebop            Cowboy Bebop   TV  Original   \n",
       "2        6                Trigun                  Trigun   TV     Manga   \n",
       "3        7    Witch Hunter Robin      Witch Hunter Robin   TV  Original   \n",
       "4        8        Bouken Ou Beet  Beet the Vandel Buster   TV     Manga   \n",
       "5       16  Hachimitsu to Clover        Honey and Clover   TV     Manga   \n",
       "\n",
       "                                               genre  episodes  score  \\\n",
       "0    Action, Adventure, Comedy, Drama, Sci-Fi, Space      26.0   8.81   \n",
       "2  Action, Sci-Fi, Adventure, Comedy, Drama, Shounen      26.0   8.30   \n",
       "3  Action, Magic, Police, Supernatural, Drama, My...      26.0   7.33   \n",
       "4          Adventure, Fantasy, Shounen, Supernatural      52.0   7.03   \n",
       "5       Comedy, Drama, Josei, Romance, Slice of Life      24.0   8.12   \n",
       "\n",
       "   scored_by  \n",
       "0   405664.0  \n",
       "2   212537.0  \n",
       "3    32837.0  \n",
       "4     4894.0  \n",
       "5    57065.0  "
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tv.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Movies have 1 episode\n",
    "movie = anime_data[anime_data['type'] == 'Movie']\n",
    "\n",
    "#filling Movies episode null value with 1\n",
    "movie['episodes'].fillna(1, inplace = True)\n",
    "\n",
    "#finding average movie rating and average scored by\n",
    "movie_rating_avg = movie['score'].mean(skipna=True)\n",
    "movie_scoredBy_avg = math.floor(movie['scored_by'].mean(skipna=True))\n",
    "\n",
    "#filling empty movie rating and scored by with average\n",
    "movie['score'].fillna(movie_rating_avg, inplace = True)\n",
    "movie['scored_by'].fillna(movie_scoredBy_avg, inplace = True)\n",
    "\n",
    "#merging these values in main anime_data\n",
    "anime_data.update(movie)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "#finding median of OVA  episodes\n",
    "ova = anime_data[anime_data['type'] == 'OVA']\n",
    "ova_median = ova['episodes'].median(skipna=True)\n",
    "\n",
    "#filling OVA episodes null value with median\n",
    "ova['episodes'].fillna(ova_median, inplace = True)\n",
    "\n",
    "#finding average OVA rating and average scored by\n",
    "ova_rating_avg = ova['score'].mean(skipna=True)\n",
    "ova_scoredBy_avg = math.floor(ova['scored_by'].mean(skipna=True))\n",
    "\n",
    "#filling empty OVA rating and scored by with average\n",
    "ova['score'].fillna(ova_rating_avg, inplace = True)\n",
    "ova['scored_by'].fillna(ova_scoredBy_avg, inplace = True)\n",
    "\n",
    "#merging these values in main anime_data\n",
    "anime_data.update(ova)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "#finding median of Special episodes\n",
    "special = anime_data[anime_data['type'] == 'Special']\n",
    "special_median = special['episodes'].median(skipna=True)\n",
    "\n",
    "#filling Special episodes null value with median\n",
    "special['episodes'].fillna(special_median, inplace = True)\n",
    "\n",
    "#finding average Special rating and average scored by\n",
    "special_rating_avg = special['score'].mean(skipna=True)\n",
    "special_scoredBy_avg = math.floor(special['scored_by'].mean(skipna=True))\n",
    "\n",
    "#filling empty Special rating and scored by with average\n",
    "special['score'].fillna(special_rating_avg, inplace = True)\n",
    "special['scored_by'].fillna(special_scoredBy_avg, inplace = True)\n",
    "\n",
    "#merging these values in main anime_data\n",
    "anime_data.update(special)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "#finding median of ONA  episodes\n",
    "ona = anime_data[anime_data['type'] == 'ONA']\n",
    "ona_median = ona['episodes'].median(skipna=True)\n",
    "\n",
    "#filling ONA null value with median\n",
    "ona['episodes'].fillna(ona_median, inplace = True)\n",
    "\n",
    "#finding average ONA rating and average scored by\n",
    "ona_rating_avg = ona['score'].mean(skipna=True)\n",
    "ona_scoredBy_avg = math.floor(ona['scored_by'].mean(skipna=True))\n",
    "\n",
    "#filling empty ONA rating and scored by with average\n",
    "ona['score'].fillna(ona_rating_avg, inplace = True)\n",
    "ona['scored_by'].fillna(ona_scoredBy_avg, inplace = True)\n",
    "\n",
    "#mergin these values in main anime_data\n",
    "anime_data.update(ona)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "#finding median of Music  episodes\n",
    "music = anime_data[anime_data['type'] == 'Music']\n",
    "music_median = ova['episodes'].median(skipna=True)\n",
    "\n",
    "#filling Music null value with median\n",
    "music['episodes'].fillna(music_median, inplace = True)\n",
    "\n",
    "#finding average Music rating and average scored by\n",
    "music_rating_avg = music['score'].mean(skipna=True)\n",
    "music_scoredBy_avg = math.floor(music['scored_by'].mean(skipna=True))\n",
    "\n",
    "#filling empty Music rating and scored by with average\n",
    "music['score'].fillna(music_rating_avg, inplace = True)\n",
    "music['scored_by'].fillna(music_scoredBy_avg, inplace = True)\n",
    "\n",
    "#mergin these values in main anime_data\n",
    "anime_data.update(music)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2.0"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#checking if a Unknown values are replaced\n",
    "episode = anime_data['episodes']\n",
    "episode[15135]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "#nan_rows = anime_data[anime_data.isnull().any(1)]\n",
    "#nan_rows.head(50)\n",
    "#anime_data.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "finding weighted rating"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "#finding mean rating across the whole anime_data\n",
    "mean_animeRating = anime_data['score'].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "#funtion to find weighted reating\n",
    "def W_R(a, C=mean_animeRating):\n",
    "    v = a['scored_by']\n",
    "    R = a['score']\n",
    "    return (v/(v+1) * R) + (1/(1+v) * C)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "#new column with weighted rating\n",
    "anime_data['rating'] = anime_data.apply(W_R, axis=1)\n",
    "#anime_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "#anime_data.to_csv(r'D:\\NSU\\11th Semester\\CSE445 - AZK\\Project\\test1.csv', index=False, mode= 'w')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "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>animeID</th>\n",
       "      <th>name</th>\n",
       "      <th>title_english</th>\n",
       "      <th>type</th>\n",
       "      <th>source</th>\n",
       "      <th>genre</th>\n",
       "      <th>episodes</th>\n",
       "      <th>score</th>\n",
       "      <th>scored_by</th>\n",
       "      <th>rating</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Cowboy Bebop</td>\n",
       "      <td>Cowboy Bebop</td>\n",
       "      <td>TV</td>\n",
       "      <td>Original</td>\n",
       "      <td>Action, Adventure, Comedy, Drama, Sci-Fi, Space</td>\n",
       "      <td>26.0</td>\n",
       "      <td>8.81</td>\n",
       "      <td>405664.0</td>\n",
       "      <td>8.809994</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>5.0</td>\n",
       "      <td>Cowboy Bebop: Tengoku no Tobira</td>\n",
       "      <td>Cowboy Bebop: The Movie</td>\n",
       "      <td>Movie</td>\n",
       "      <td>Original</td>\n",
       "      <td>Action, Drama, Mystery, Sci-Fi, Space</td>\n",
       "      <td>1.0</td>\n",
       "      <td>8.41</td>\n",
       "      <td>120243.0</td>\n",
       "      <td>8.409982</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>6.0</td>\n",
       "      <td>Trigun</td>\n",
       "      <td>Trigun</td>\n",
       "      <td>TV</td>\n",
       "      <td>Manga</td>\n",
       "      <td>Action, Sci-Fi, Adventure, Comedy, Drama, Shounen</td>\n",
       "      <td>26.0</td>\n",
       "      <td>8.30</td>\n",
       "      <td>212537.0</td>\n",
       "      <td>8.299991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>7.0</td>\n",
       "      <td>Witch Hunter Robin</td>\n",
       "      <td>Witch Hunter Robin</td>\n",
       "      <td>TV</td>\n",
       "      <td>Original</td>\n",
       "      <td>Action, Magic, Police, Supernatural, Drama, My...</td>\n",
       "      <td>26.0</td>\n",
       "      <td>7.33</td>\n",
       "      <td>32837.0</td>\n",
       "      <td>7.329968</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>8.0</td>\n",
       "      <td>Bouken Ou Beet</td>\n",
       "      <td>Beet the Vandel Buster</td>\n",
       "      <td>TV</td>\n",
       "      <td>Manga</td>\n",
       "      <td>Adventure, Fantasy, Shounen, Supernatural</td>\n",
       "      <td>52.0</td>\n",
       "      <td>7.03</td>\n",
       "      <td>4894.0</td>\n",
       "      <td>7.029848</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   animeID                             name            title_english   type  \\\n",
       "0      1.0                     Cowboy Bebop             Cowboy Bebop     TV   \n",
       "1      5.0  Cowboy Bebop: Tengoku no Tobira  Cowboy Bebop: The Movie  Movie   \n",
       "2      6.0                           Trigun                   Trigun     TV   \n",
       "3      7.0               Witch Hunter Robin       Witch Hunter Robin     TV   \n",
       "4      8.0                   Bouken Ou Beet   Beet the Vandel Buster     TV   \n",
       "\n",
       "     source                                              genre  episodes  \\\n",
       "0  Original    Action, Adventure, Comedy, Drama, Sci-Fi, Space      26.0   \n",
       "1  Original              Action, Drama, Mystery, Sci-Fi, Space       1.0   \n",
       "2     Manga  Action, Sci-Fi, Adventure, Comedy, Drama, Shounen      26.0   \n",
       "3  Original  Action, Magic, Police, Supernatural, Drama, My...      26.0   \n",
       "4     Manga          Adventure, Fantasy, Shounen, Supernatural      52.0   \n",
       "\n",
       "   score  scored_by    rating  \n",
       "0   8.81   405664.0  8.809994  \n",
       "1   8.41   120243.0  8.409982  \n",
       "2   8.30   212537.0  8.299991  \n",
       "3   7.33    32837.0  7.329968  \n",
       "4   7.03     4894.0  7.029848  "
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "anime_data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "User Data file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "#reading user data csv\n",
    "user_data = pd.read_csv(path2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "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>profile</th>\n",
       "      <th>gender</th>\n",
       "      <th>birthday</th>\n",
       "      <th>favorites_anime</th>\n",
       "      <th>link</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>DesolatePsyche</td>\n",
       "      <td>Male</td>\n",
       "      <td>02-Oct-94</td>\n",
       "      <td>['33352', '25013', '5530', '33674', '1482', '2...</td>\n",
       "      <td>https://myanimelist.net/profile/DesolatePsyche</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>baekbeans</td>\n",
       "      <td>Female</td>\n",
       "      <td>10-Nov-00</td>\n",
       "      <td>['11061', '31964', '853', '20583', '918', '925...</td>\n",
       "      <td>https://myanimelist.net/profile/baekbeans</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>skrn</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>['918', '2904', '11741', '17074', '23273', '32...</td>\n",
       "      <td>https://myanimelist.net/profile/skrn</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>edgewalker00</td>\n",
       "      <td>Male</td>\n",
       "      <td>Sep-05</td>\n",
       "      <td>['5680', '849', '2904', '3588', '37349']</td>\n",
       "      <td>https://myanimelist.net/profile/edgewalker00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>aManOfCulture99</td>\n",
       "      <td>Male</td>\n",
       "      <td>30-Oct-99</td>\n",
       "      <td>['4181', '7791', '9617', '5680', '2167', '4382...</td>\n",
       "      <td>https://myanimelist.net/profile/aManOfCulture99</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           profile  gender   birthday  \\\n",
       "0   DesolatePsyche    Male  02-Oct-94   \n",
       "1        baekbeans  Female  10-Nov-00   \n",
       "2             skrn     NaN        NaN   \n",
       "3     edgewalker00    Male     Sep-05   \n",
       "4  aManOfCulture99    Male  30-Oct-99   \n",
       "\n",
       "                                     favorites_anime  \\\n",
       "0  ['33352', '25013', '5530', '33674', '1482', '2...   \n",
       "1  ['11061', '31964', '853', '20583', '918', '925...   \n",
       "2  ['918', '2904', '11741', '17074', '23273', '32...   \n",
       "3           ['5680', '849', '2904', '3588', '37349']   \n",
       "4  ['4181', '7791', '9617', '5680', '2167', '4382...   \n",
       "\n",
       "                                              link  \n",
       "0   https://myanimelist.net/profile/DesolatePsyche  \n",
       "1        https://myanimelist.net/profile/baekbeans  \n",
       "2             https://myanimelist.net/profile/skrn  \n",
       "3     https://myanimelist.net/profile/edgewalker00  \n",
       "4  https://myanimelist.net/profile/aManOfCulture99  "
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "user_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "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>profile</th>\n",
       "      <th>gender</th>\n",
       "      <th>birthday</th>\n",
       "      <th>favorites_anime</th>\n",
       "      <th>link</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>DesolatePsyche</td>\n",
       "      <td>Male</td>\n",
       "      <td>02-Oct-94</td>\n",
       "      <td>['33352', '25013', '5530', '33674', '1482', '2...</td>\n",
       "      <td>https://myanimelist.net/profile/DesolatePsyche</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>baekbeans</td>\n",
       "      <td>Female</td>\n",
       "      <td>10-Nov-00</td>\n",
       "      <td>['11061', '31964', '853', '20583', '918', '925...</td>\n",
       "      <td>https://myanimelist.net/profile/baekbeans</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>skrn</td>\n",
       "      <td>unknown</td>\n",
       "      <td>NaN</td>\n",
       "      <td>['918', '2904', '11741', '17074', '23273', '32...</td>\n",
       "      <td>https://myanimelist.net/profile/skrn</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>edgewalker00</td>\n",
       "      <td>Male</td>\n",
       "      <td>Sep-05</td>\n",
       "      <td>['5680', '849', '2904', '3588', '37349']</td>\n",
       "      <td>https://myanimelist.net/profile/edgewalker00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>aManOfCulture99</td>\n",
       "      <td>Male</td>\n",
       "      <td>30-Oct-99</td>\n",
       "      <td>['4181', '7791', '9617', '5680', '2167', '4382...</td>\n",
       "      <td>https://myanimelist.net/profile/aManOfCulture99</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           profile   gender   birthday  \\\n",
       "0   DesolatePsyche     Male  02-Oct-94   \n",
       "1        baekbeans   Female  10-Nov-00   \n",
       "2             skrn  unknown        NaN   \n",
       "3     edgewalker00     Male     Sep-05   \n",
       "4  aManOfCulture99     Male  30-Oct-99   \n",
       "\n",
       "                                     favorites_anime  \\\n",
       "0  ['33352', '25013', '5530', '33674', '1482', '2...   \n",
       "1  ['11061', '31964', '853', '20583', '918', '925...   \n",
       "2  ['918', '2904', '11741', '17074', '23273', '32...   \n",
       "3           ['5680', '849', '2904', '3588', '37349']   \n",
       "4  ['4181', '7791', '9617', '5680', '2167', '4382...   \n",
       "\n",
       "                                              link  \n",
       "0   https://myanimelist.net/profile/DesolatePsyche  \n",
       "1        https://myanimelist.net/profile/baekbeans  \n",
       "2             https://myanimelist.net/profile/skrn  \n",
       "3     https://myanimelist.net/profile/edgewalker00  \n",
       "4  https://myanimelist.net/profile/aManOfCulture99  "
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#replacing empty cells as unknown gender\n",
    "user_data['gender'].fillna('unknown', inplace = True)\n",
    "user_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "#removed unknown type gender\n",
    "user_data = user_data[user_data.gender != 'unknown']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "53856"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(user_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    33352, 25013, 5530, 33674, 1482, 269, 18245, 2...\n",
       "1    11061, 31964, 853, 20583, 918, 9253, 34599, 35...\n",
       "3                         5680, 849, 2904, 3588, 37349\n",
       "4    4181, 7791, 9617, 5680, 2167, 4382, 849, 235, ...\n",
       "7    11061, 30, 6594, 28701, 10087, 6746, 918, 1535...\n",
       "Name: favorites_anime, dtype: object"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#replacing [,] and ' from the favourites anime\n",
    "temp = user_data['favorites_anime']\n",
    "\n",
    "temp = temp.str.replace('[','',regex=True)\n",
    "temp = temp.str.replace(']','',regex=True)\n",
    "temp = temp.str.replace(\"'\",'',regex=True)\n",
    "temp.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "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>profile</th>\n",
       "      <th>gender</th>\n",
       "      <th>birthday</th>\n",
       "      <th>favorites_anime</th>\n",
       "      <th>link</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>DesolatePsyche</td>\n",
       "      <td>Male</td>\n",
       "      <td>02-Oct-94</td>\n",
       "      <td>33352, 25013, 5530, 33674, 1482, 269, 18245, 2...</td>\n",
       "      <td>https://myanimelist.net/profile/DesolatePsyche</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>baekbeans</td>\n",
       "      <td>Female</td>\n",
       "      <td>10-Nov-00</td>\n",
       "      <td>11061, 31964, 853, 20583, 918, 9253, 34599, 35...</td>\n",
       "      <td>https://myanimelist.net/profile/baekbeans</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>edgewalker00</td>\n",
       "      <td>Male</td>\n",
       "      <td>Sep-05</td>\n",
       "      <td>5680, 849, 2904, 3588, 37349</td>\n",
       "      <td>https://myanimelist.net/profile/edgewalker00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>aManOfCulture99</td>\n",
       "      <td>Male</td>\n",
       "      <td>30-Oct-99</td>\n",
       "      <td>4181, 7791, 9617, 5680, 2167, 4382, 849, 235, ...</td>\n",
       "      <td>https://myanimelist.net/profile/aManOfCulture99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>NIGGER_BONER</td>\n",
       "      <td>Male</td>\n",
       "      <td>01-Jan-85</td>\n",
       "      <td>11061, 30, 6594, 28701, 10087, 6746, 918, 1535...</td>\n",
       "      <td>https://myanimelist.net/profile/NIGGER_BONER</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           profile  gender   birthday  \\\n",
       "0   DesolatePsyche    Male  02-Oct-94   \n",
       "1        baekbeans  Female  10-Nov-00   \n",
       "3     edgewalker00    Male     Sep-05   \n",
       "4  aManOfCulture99    Male  30-Oct-99   \n",
       "7     NIGGER_BONER    Male  01-Jan-85   \n",
       "\n",
       "                                     favorites_anime  \\\n",
       "0  33352, 25013, 5530, 33674, 1482, 269, 18245, 2...   \n",
       "1  11061, 31964, 853, 20583, 918, 9253, 34599, 35...   \n",
       "3                       5680, 849, 2904, 3588, 37349   \n",
       "4  4181, 7791, 9617, 5680, 2167, 4382, 849, 235, ...   \n",
       "7  11061, 30, 6594, 28701, 10087, 6746, 918, 1535...   \n",
       "\n",
       "                                              link  \n",
       "0   https://myanimelist.net/profile/DesolatePsyche  \n",
       "1        https://myanimelist.net/profile/baekbeans  \n",
       "3     https://myanimelist.net/profile/edgewalker00  \n",
       "4  https://myanimelist.net/profile/aManOfCulture99  \n",
       "7     https://myanimelist.net/profile/NIGGER_BONER  "
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#updating favourites anime\n",
    "user_data['favorites_anime'] = temp\n",
    "user_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "#separating favourite anime genderwise\n",
    "male_fav=user_data[user_data['gender']=='Male']\n",
    "female_fav=user_data[user_data['gender']=='Female']\n",
    "#male_fav.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "#counting female fans of anime id wise\n",
    "f_data = female_fav.favorites_anime.str.get_dummies(sep=', ').sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1        597\n",
       "100       27\n",
       "1000       6\n",
       "10012     12\n",
       "10015      4\n",
       "        ... \n",
       "9981       3\n",
       "9982       5\n",
       "9989     333\n",
       "9996       2\n",
       "9999       2\n",
       "Length: 2633, dtype: int64"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "#counting male fans of anime id wise\n",
    "m_data = male_fav.favorites_anime.str.get_dummies(sep=', ').sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "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>animeID</th>\n",
       "      <th>female_fan</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>597</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>100</td>\n",
       "      <td>27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>1000</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>10012</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>10015</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  animeID  female_fan\n",
       "0       1         597\n",
       "1     100          27\n",
       "2    1000           6\n",
       "3   10012          12\n",
       "4   10015           4"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#creating a dataframe with anime id and female fan number\n",
    "temp_1 = pd.DataFrame({'animeID': f_data.index, 'female_fan': f_data})\n",
    "#df = pd.DataFrame({'female_fan': f_data})\n",
    "temp_1.reset_index(drop=True, inplace=True)\n",
    "temp_1.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "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>animeID</th>\n",
       "      <th>male_fan</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2922</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>100</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>1000</td>\n",
       "      <td>23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>10012</td>\n",
       "      <td>67</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>10015</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  animeID  male_fan\n",
       "0       1      2922\n",
       "1     100         1\n",
       "2    1000        23\n",
       "3   10012        67\n",
       "4   10015         2"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#creating another dataframe with anime id and male fan number\n",
    "temp_2 = pd.DataFrame({'animeID': m_data.index, 'male_fan': m_data})\n",
    "temp_2.reset_index(drop=True, inplace=True)\n",
    "temp_2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "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>animeID</th>\n",
       "      <th>male_fan</th>\n",
       "      <th>female_fan</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2922</td>\n",
       "      <td>597.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>100</td>\n",
       "      <td>1</td>\n",
       "      <td>27.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>1000</td>\n",
       "      <td>23</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>10012</td>\n",
       "      <td>67</td>\n",
       "      <td>12.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>10015</td>\n",
       "      <td>2</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  animeID  male_fan  female_fan\n",
       "0       1      2922       597.0\n",
       "1     100         1        27.0\n",
       "2    1000        23         6.0\n",
       "3   10012        67        12.0\n",
       "4   10015         2         4.0"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#merging two dataframe based on anime id\n",
    "#temp_3 =  pd.DataFrame.merge(temp_2,temp_1,on='animeID')\n",
    "temp_3 = temp_2.merge(temp_1,how='left', left_on='animeID', right_on='animeID')\n",
    "temp_3.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "#temp_3.to_csv(r'D:\\NSU\\11th Semester\\CSE445 - AZK\\Project\\user_data.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "#converting anime id from str to numeric\n",
    "temp_3[\"animeID\"] = pd.to_numeric(temp_3[\"animeID\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "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>animeID</th>\n",
       "      <th>name</th>\n",
       "      <th>title_english</th>\n",
       "      <th>type</th>\n",
       "      <th>source</th>\n",
       "      <th>genre</th>\n",
       "      <th>episodes</th>\n",
       "      <th>score</th>\n",
       "      <th>scored_by</th>\n",
       "      <th>rating</th>\n",
       "      <th>male_fan</th>\n",
       "      <th>female_fan</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Cowboy Bebop</td>\n",
       "      <td>Cowboy Bebop</td>\n",
       "      <td>TV</td>\n",
       "      <td>Original</td>\n",
       "      <td>Action, Adventure, Comedy, Drama, Sci-Fi, Space</td>\n",
       "      <td>26.0</td>\n",
       "      <td>8.81</td>\n",
       "      <td>405664.0</td>\n",
       "      <td>8.809994</td>\n",
       "      <td>2922.0</td>\n",
       "      <td>597.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>5.0</td>\n",
       "      <td>Cowboy Bebop: Tengoku no Tobira</td>\n",
       "      <td>Cowboy Bebop: The Movie</td>\n",
       "      <td>Movie</td>\n",
       "      <td>Original</td>\n",
       "      <td>Action, Drama, Mystery, Sci-Fi, Space</td>\n",
       "      <td>1.0</td>\n",
       "      <td>8.41</td>\n",
       "      <td>120243.0</td>\n",
       "      <td>8.409982</td>\n",
       "      <td>36.0</td>\n",
       "      <td>11.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>6.0</td>\n",
       "      <td>Trigun</td>\n",
       "      <td>Trigun</td>\n",
       "      <td>TV</td>\n",
       "      <td>Manga</td>\n",
       "      <td>Action, Sci-Fi, Adventure, Comedy, Drama, Shounen</td>\n",
       "      <td>26.0</td>\n",
       "      <td>8.30</td>\n",
       "      <td>212537.0</td>\n",
       "      <td>8.299991</td>\n",
       "      <td>609.0</td>\n",
       "      <td>179.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>7.0</td>\n",
       "      <td>Witch Hunter Robin</td>\n",
       "      <td>Witch Hunter Robin</td>\n",
       "      <td>TV</td>\n",
       "      <td>Original</td>\n",
       "      <td>Action, Magic, Police, Supernatural, Drama, My...</td>\n",
       "      <td>26.0</td>\n",
       "      <td>7.33</td>\n",
       "      <td>32837.0</td>\n",
       "      <td>7.329968</td>\n",
       "      <td>18.0</td>\n",
       "      <td>24.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>8.0</td>\n",
       "      <td>Bouken Ou Beet</td>\n",
       "      <td>Beet the Vandel Buster</td>\n",
       "      <td>TV</td>\n",
       "      <td>Manga</td>\n",
       "      <td>Adventure, Fantasy, Shounen, Supernatural</td>\n",
       "      <td>52.0</td>\n",
       "      <td>7.03</td>\n",
       "      <td>4894.0</td>\n",
       "      <td>7.029848</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   animeID                             name            title_english   type  \\\n",
       "0      1.0                     Cowboy Bebop             Cowboy Bebop     TV   \n",
       "1      5.0  Cowboy Bebop: Tengoku no Tobira  Cowboy Bebop: The Movie  Movie   \n",
       "2      6.0                           Trigun                   Trigun     TV   \n",
       "3      7.0               Witch Hunter Robin       Witch Hunter Robin     TV   \n",
       "4      8.0                   Bouken Ou Beet   Beet the Vandel Buster     TV   \n",
       "\n",
       "     source                                              genre  episodes  \\\n",
       "0  Original    Action, Adventure, Comedy, Drama, Sci-Fi, Space      26.0   \n",
       "1  Original              Action, Drama, Mystery, Sci-Fi, Space       1.0   \n",
       "2     Manga  Action, Sci-Fi, Adventure, Comedy, Drama, Shounen      26.0   \n",
       "3  Original  Action, Magic, Police, Supernatural, Drama, My...      26.0   \n",
       "4     Manga          Adventure, Fantasy, Shounen, Supernatural      52.0   \n",
       "\n",
       "   score  scored_by    rating  male_fan  female_fan  \n",
       "0   8.81   405664.0  8.809994    2922.0       597.0  \n",
       "1   8.41   120243.0  8.409982      36.0        11.0  \n",
       "2   8.30   212537.0  8.299991     609.0       179.0  \n",
       "3   7.33    32837.0  7.329968      18.0        24.0  \n",
       "4   7.03     4894.0  7.029848       NaN         NaN  "
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#creating a new dataframe combining anime data and user data\n",
    "#df =  pd.DataFrame.merge(anime_data,temp_3,on='animeID')\n",
    "df = anime_data.merge(temp_3,how='left', left_on='animeID', right_on='animeID')\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "#filling empty cells with 0\n",
    "df['male_fan'].fillna(0, inplace=True)\n",
    "df['female_fan'].fillna(0, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "#funtion to calculate fanbase gender biasness\n",
    "def Fan(a):\n",
    "    x = a['male_fan']\n",
    "    y = a['female_fan']\n",
    "    return (x-y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "#new column with fan base\n",
    "df['fan_base'] = df.apply(Fan, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "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>animeID</th>\n",
       "      <th>name</th>\n",
       "      <th>title_english</th>\n",
       "      <th>type</th>\n",
       "      <th>source</th>\n",
       "      <th>genre</th>\n",
       "      <th>episodes</th>\n",
       "      <th>score</th>\n",
       "      <th>scored_by</th>\n",
       "      <th>rating</th>\n",
       "      <th>male_fan</th>\n",
       "      <th>female_fan</th>\n",
       "      <th>fan_base</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Cowboy Bebop</td>\n",
       "      <td>Cowboy Bebop</td>\n",
       "      <td>TV</td>\n",
       "      <td>Original</td>\n",
       "      <td>Action, Adventure, Comedy, Drama, Sci-Fi, Space</td>\n",
       "      <td>26.0</td>\n",
       "      <td>8.81</td>\n",
       "      <td>405664.0</td>\n",
       "      <td>8.809994</td>\n",
       "      <td>2922.0</td>\n",
       "      <td>597.0</td>\n",
       "      <td>2325.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>5.0</td>\n",
       "      <td>Cowboy Bebop: Tengoku no Tobira</td>\n",
       "      <td>Cowboy Bebop: The Movie</td>\n",
       "      <td>Movie</td>\n",
       "      <td>Original</td>\n",
       "      <td>Action, Drama, Mystery, Sci-Fi, Space</td>\n",
       "      <td>1.0</td>\n",
       "      <td>8.41</td>\n",
       "      <td>120243.0</td>\n",
       "      <td>8.409982</td>\n",
       "      <td>36.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>25.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>6.0</td>\n",
       "      <td>Trigun</td>\n",
       "      <td>Trigun</td>\n",
       "      <td>TV</td>\n",
       "      <td>Manga</td>\n",
       "      <td>Action, Sci-Fi, Adventure, Comedy, Drama, Shounen</td>\n",
       "      <td>26.0</td>\n",
       "      <td>8.30</td>\n",
       "      <td>212537.0</td>\n",
       "      <td>8.299991</td>\n",
       "      <td>609.0</td>\n",
       "      <td>179.0</td>\n",
       "      <td>430.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>7.0</td>\n",
       "      <td>Witch Hunter Robin</td>\n",
       "      <td>Witch Hunter Robin</td>\n",
       "      <td>TV</td>\n",
       "      <td>Original</td>\n",
       "      <td>Action, Magic, Police, Supernatural, Drama, My...</td>\n",
       "      <td>26.0</td>\n",
       "      <td>7.33</td>\n",
       "      <td>32837.0</td>\n",
       "      <td>7.329968</td>\n",
       "      <td>18.0</td>\n",
       "      <td>24.0</td>\n",
       "      <td>-6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>8.0</td>\n",
       "      <td>Bouken Ou Beet</td>\n",
       "      <td>Beet the Vandel Buster</td>\n",
       "      <td>TV</td>\n",
       "      <td>Manga</td>\n",
       "      <td>Adventure, Fantasy, Shounen, Supernatural</td>\n",
       "      <td>52.0</td>\n",
       "      <td>7.03</td>\n",
       "      <td>4894.0</td>\n",
       "      <td>7.029848</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   animeID                             name            title_english   type  \\\n",
       "0      1.0                     Cowboy Bebop             Cowboy Bebop     TV   \n",
       "1      5.0  Cowboy Bebop: Tengoku no Tobira  Cowboy Bebop: The Movie  Movie   \n",
       "2      6.0                           Trigun                   Trigun     TV   \n",
       "3      7.0               Witch Hunter Robin       Witch Hunter Robin     TV   \n",
       "4      8.0                   Bouken Ou Beet   Beet the Vandel Buster     TV   \n",
       "\n",
       "     source                                              genre  episodes  \\\n",
       "0  Original    Action, Adventure, Comedy, Drama, Sci-Fi, Space      26.0   \n",
       "1  Original              Action, Drama, Mystery, Sci-Fi, Space       1.0   \n",
       "2     Manga  Action, Sci-Fi, Adventure, Comedy, Drama, Shounen      26.0   \n",
       "3  Original  Action, Magic, Police, Supernatural, Drama, My...      26.0   \n",
       "4     Manga          Adventure, Fantasy, Shounen, Supernatural      52.0   \n",
       "\n",
       "   score  scored_by    rating  male_fan  female_fan  fan_base  \n",
       "0   8.81   405664.0  8.809994    2922.0       597.0    2325.0  \n",
       "1   8.41   120243.0  8.409982      36.0        11.0      25.0  \n",
       "2   8.30   212537.0  8.299991     609.0       179.0     430.0  \n",
       "3   7.33    32837.0  7.329968      18.0        24.0      -6.0  \n",
       "4   7.03     4894.0  7.029848       0.0         0.0       0.0  "
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "#saving dataframe to csv file\n",
    "df.to_csv(r'D:\\NSU\\11th Semester\\CSE445 - AZK\\Project\\anime_data.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv(r'D:\\NSU\\11th Semester\\CSE445 - AZK\\Project\\anime_data.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "#converting feature to categorical values\n",
    "temp_type = pd.get_dummies(df['type'])\n",
    "temp_source = pd.get_dummies(df['source'])\n",
    "temp_genre = df['genre'].str.get_dummies(sep=', ')\n",
    "\n",
    "#temp_type.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "#taking the features into a dataframe\n",
    "anime = pd.concat([temp_type,temp_source,temp_genre,df['episodes'],df['rating'],df['fan_base']], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "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>Movie</th>\n",
       "      <th>Music</th>\n",
       "      <th>ONA</th>\n",
       "      <th>OVA</th>\n",
       "      <th>Special</th>\n",
       "      <th>TV</th>\n",
       "      <th>4-koma manga</th>\n",
       "      <th>Book</th>\n",
       "      <th>Card game</th>\n",
       "      <th>Digital manga</th>\n",
       "      <th>...</th>\n",
       "      <th>Sports</th>\n",
       "      <th>Super Power</th>\n",
       "      <th>Supernatural</th>\n",
       "      <th>Thriller</th>\n",
       "      <th>Vampire</th>\n",
       "      <th>Yaoi</th>\n",
       "      <th>Yuri</th>\n",
       "      <th>episodes</th>\n",
       "      <th>rating</th>\n",
       "      <th>fan_base</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\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",
       "      <td>0</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>26.0</td>\n",
       "      <td>8.809994</td>\n",
       "      <td>2325.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</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>8.409982</td>\n",
       "      <td>25.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</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",
       "      <td>0</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>26.0</td>\n",
       "      <td>8.299991</td>\n",
       "      <td>430.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</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",
       "      <td>0</td>\n",
       "      <td>...</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>26.0</td>\n",
       "      <td>7.329968</td>\n",
       "      <td>-6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</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",
       "      <td>0</td>\n",
       "      <td>...</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>52.0</td>\n",
       "      <td>7.029848</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 67 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Movie  Music  ONA  OVA  Special  TV  4-koma manga  Book  Card game  \\\n",
       "0      0      0    0    0        0   1             0     0          0   \n",
       "1      1      0    0    0        0   0             0     0          0   \n",
       "2      0      0    0    0        0   1             0     0          0   \n",
       "3      0      0    0    0        0   1             0     0          0   \n",
       "4      0      0    0    0        0   1             0     0          0   \n",
       "\n",
       "   Digital manga  ...  Sports  Super Power  Supernatural  Thriller  Vampire  \\\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             1         0        0   \n",
       "4              0  ...       0            0             1         0        0   \n",
       "\n",
       "   Yaoi  Yuri  episodes    rating  fan_base  \n",
       "0     0     0      26.0  8.809994    2325.0  \n",
       "1     0     0       1.0  8.409982      25.0  \n",
       "2     0     0      26.0  8.299991     430.0  \n",
       "3     0     0      26.0  7.329968      -6.0  \n",
       "4     0     0      52.0  7.029848       0.0  \n",
       "\n",
       "[5 rows x 67 columns]"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "anime.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1880cfc62c8>"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1296 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#correlation for features\n",
    "# Compute the correlation matrix\n",
    "corr = anime.corr()\n",
    "\n",
    "# Generate a mask for the upper triangle\n",
    "#mask = np.triu(np.ones_like(corr, dtype=bool))\n",
    "\n",
    "# Set up the matplotlib figure\n",
    "f, ax = plt.subplots(figsize=(20, 18))\n",
    "\n",
    "# Generate a custom diverging colormap\n",
    "cmap = sns.diverging_palette(230, 20, as_cmap=True)\n",
    "\n",
    "# Draw the heatmap with the mask and correct aspect ratio\n",
    "sns.heatmap(corr, cmap=cmap, center=0,\n",
    "            square=True, linewidths=.5, cbar_kws={\"shrink\": .5})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.  , 0.  , 0.  , ..., 0.01, 0.94, 0.58],\n",
       "       [1.  , 0.  , 0.  , ..., 0.  , 0.89, 0.18],\n",
       "       [0.  , 0.  , 0.  , ..., 0.01, 0.87, 0.25],\n",
       "       ...,\n",
       "       [0.  , 0.  , 1.  , ..., 0.  , 0.52, 0.17],\n",
       "       [0.  , 0.  , 0.  , ..., 0.01, 0.66, 0.18],\n",
       "       [0.  , 0.  , 0.  , ..., 0.01, 0.66, 0.17]])"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Scaling the values\n",
    "scaler = MinMaxScaler()\n",
    "anime = scaler.fit_transform(anime)\n",
    "np.round(anime,2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Unsupervised K Nearest Neighbors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "#clustering using KNN\n",
    "nbrs = NearestNeighbors(n_neighbors=11, algorithm='ball_tree').fit(anime)\n",
    "distances, indices = nbrs.kneighbors(anime)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "#funtion to get recommendation\n",
    "def get_recommendation(name):\n",
    "    rec_id=df.index[df['name']==name].tolist()[0]\n",
    "    recommend = pd.DataFrame(columns = ['Recommended Anime', 'English Title', 'Type', 'Rating', 'Genre'])\n",
    "    for id in indices[rec_id][1:]:\n",
    "        new_row = {'Recommended Anime':df.iloc[id]['name'], 'English Title':df.iloc[id]['title_english'], 'Type':df.iloc[id]['type'], 'Rating':np.round(df.iloc[id]['rating'],2), 'Genre':df.iloc[id]['genre'] }\n",
    "        recommend = recommend.append(new_row, ignore_index=True)\n",
    "    print(\"Anime Recommendation for:\", name, '/',df.iloc[rec_id]['title_english'])\n",
    "    display(recommend)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Getting Recommendation using KNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Anime Recommendation for: Cowboy Bebop / Cowboy Bebop\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>Recommended Anime</th>\n",
       "      <th>English Title</th>\n",
       "      <th>Type</th>\n",
       "      <th>Rating</th>\n",
       "      <th>Genre</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>Ginga Tetsudou Monogatari</td>\n",
       "      <td>The Galaxy Railways</td>\n",
       "      <td>TV</td>\n",
       "      <td>7.25</td>\n",
       "      <td>Action, Adventure, Space, Drama, Sci-Fi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>Ginga Tetsudou Monogatari: Eien e no Bunkiten</td>\n",
       "      <td>NaN</td>\n",
       "      <td>TV</td>\n",
       "      <td>6.98</td>\n",
       "      <td>Action, Adventure, Space, Drama, Sci-Fi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>Seihou Tenshi Angel Links</td>\n",
       "      <td>Angel Links</td>\n",
       "      <td>TV</td>\n",
       "      <td>6.02</td>\n",
       "      <td>Action, Adventure, Space, Comedy, Romance, Dra...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>Uchuu Patrol Luluco</td>\n",
       "      <td>Space Patrol Luluco</td>\n",
       "      <td>TV</td>\n",
       "      <td>7.68</td>\n",
       "      <td>Action, Adventure, Comedy, Space</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>Uchuu Senkan Yamato</td>\n",
       "      <td>Star Blazers: The Quest for Iscandar</td>\n",
       "      <td>TV</td>\n",
       "      <td>7.67</td>\n",
       "      <td>Action, Adventure, Drama, Military, Sci-Fi, Space</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>Uchuu Senkan Yamato 2</td>\n",
       "      <td>Star Blazers: The Comet Empire</td>\n",
       "      <td>TV</td>\n",
       "      <td>7.55</td>\n",
       "      <td>Action, Military, Sci-Fi, Adventure, Space, Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>Cowboy Bebop: Yose Atsume Blues</td>\n",
       "      <td>Cowboy Bebop Session XX</td>\n",
       "      <td>Special</td>\n",
       "      <td>7.49</td>\n",
       "      <td>Action, Adventure, Comedy, Drama, Sci-Fi, Space</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>Uchuu Senkan Yamato III</td>\n",
       "      <td>Star Blazers: The Bolar Wars</td>\n",
       "      <td>TV</td>\n",
       "      <td>7.21</td>\n",
       "      <td>Action, Military, Sci-Fi, Adventure, Space, Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>Uchuu Densetsu Ulysses 31</td>\n",
       "      <td>Ulysses 31</td>\n",
       "      <td>TV</td>\n",
       "      <td>7.15</td>\n",
       "      <td>Action, Adventure, Sci-Fi, Space</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>Iron Kid</td>\n",
       "      <td>Eon Kid</td>\n",
       "      <td>TV</td>\n",
       "      <td>7.06</td>\n",
       "      <td>Action, Adventure, Comedy, Sci-Fi</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                               Recommended Anime  \\\n",
       "0                      Ginga Tetsudou Monogatari   \n",
       "1  Ginga Tetsudou Monogatari: Eien e no Bunkiten   \n",
       "2                      Seihou Tenshi Angel Links   \n",
       "3                            Uchuu Patrol Luluco   \n",
       "4                            Uchuu Senkan Yamato   \n",
       "5                          Uchuu Senkan Yamato 2   \n",
       "6                Cowboy Bebop: Yose Atsume Blues   \n",
       "7                        Uchuu Senkan Yamato III   \n",
       "8                      Uchuu Densetsu Ulysses 31   \n",
       "9                                       Iron Kid   \n",
       "\n",
       "                          English Title     Type  Rating  \\\n",
       "0                   The Galaxy Railways       TV    7.25   \n",
       "1                                   NaN       TV    6.98   \n",
       "2                           Angel Links       TV    6.02   \n",
       "3                   Space Patrol Luluco       TV    7.68   \n",
       "4  Star Blazers: The Quest for Iscandar       TV    7.67   \n",
       "5        Star Blazers: The Comet Empire       TV    7.55   \n",
       "6               Cowboy Bebop Session XX  Special    7.49   \n",
       "7          Star Blazers: The Bolar Wars       TV    7.21   \n",
       "8                            Ulysses 31       TV    7.15   \n",
       "9                               Eon Kid       TV    7.06   \n",
       "\n",
       "                                               Genre  \n",
       "0            Action, Adventure, Space, Drama, Sci-Fi  \n",
       "1            Action, Adventure, Space, Drama, Sci-Fi  \n",
       "2  Action, Adventure, Space, Comedy, Romance, Dra...  \n",
       "3                   Action, Adventure, Comedy, Space  \n",
       "4  Action, Adventure, Drama, Military, Sci-Fi, Space  \n",
       "5  Action, Military, Sci-Fi, Adventure, Space, Drama  \n",
       "6    Action, Adventure, Comedy, Drama, Sci-Fi, Space  \n",
       "7  Action, Military, Sci-Fi, Adventure, Space, Drama  \n",
       "8                   Action, Adventure, Sci-Fi, Space  \n",
       "9                  Action, Adventure, Comedy, Sci-Fi  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "get_recommendation('Cowboy Bebop')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Checking other Nearest Neighbors search algorithms"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Brute algorithm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Wall time: 4.7 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "nbrs1 = NearestNeighbors(n_neighbors=10, algorithm='brute').fit(anime)\n",
    "distances1, indices1 = nbrs1.kneighbors(anime)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "#funtion to get recommendation from brute\n",
    "def get_recommendation_Brute(name):\n",
    "    rec_id=df.index[df['name']==name].tolist()[0]\n",
    "    recommend = pd.DataFrame(columns = ['Recommended Anime', 'English Title', 'Type', 'Rating', 'Genre'])\n",
    "    for id in indices1[rec_id][:]:\n",
    "        new_row = {'Recommended Anime':df.iloc[id]['name'], 'English Title':df.iloc[id]['title_english'], 'Type':df.iloc[id]['type'], 'Rating':np.round(df.iloc[id]['rating'],2), 'Genre':df.iloc[id]['genre'] }\n",
    "        recommend = recommend.append(new_row, ignore_index=True)\n",
    "    print(\"Anime Recommendation for:\", name, '/',df.iloc[rec_id]['title_english'])\n",
    "    display(recommend)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.        , 1.10067573, 1.1081936 , ..., 1.48265702, 1.48775968,\n",
       "        1.4888789 ],\n",
       "       [0.        , 1.41606606, 1.41810946, ..., 1.56845625, 1.73339718,\n",
       "        1.73379833],\n",
       "       [0.        , 1.02350008, 1.41505862, ..., 1.4188526 , 1.41959297,\n",
       "        1.42004394],\n",
       "       ...,\n",
       "       [0.        , 0.01946333, 1.00000025, ..., 1.00009246, 1.00009917,\n",
       "        1.00010261],\n",
       "       [0.        , 1.0000399 , 1.00117049, ..., 1.41599442, 1.41618116,\n",
       "        1.41685002],\n",
       "       [0.        , 0.23137357, 1.00000025, ..., 1.41421672, 1.41422498,\n",
       "        1.41423112]])"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "distances1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.100676</td>\n",
       "      <td>1.108194</td>\n",
       "      <td>1.144710</td>\n",
       "      <td>1.478284</td>\n",
       "      <td>1.479574</td>\n",
       "      <td>1.481531</td>\n",
       "      <td>1.482657</td>\n",
       "      <td>1.487760</td>\n",
       "      <td>1.488879</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.416066</td>\n",
       "      <td>1.418109</td>\n",
       "      <td>1.423655</td>\n",
       "      <td>1.426490</td>\n",
       "      <td>1.437977</td>\n",
       "      <td>1.481053</td>\n",
       "      <td>1.568456</td>\n",
       "      <td>1.733397</td>\n",
       "      <td>1.733798</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.023500</td>\n",
       "      <td>1.415059</td>\n",
       "      <td>1.416404</td>\n",
       "      <td>1.416719</td>\n",
       "      <td>1.417555</td>\n",
       "      <td>1.418584</td>\n",
       "      <td>1.418853</td>\n",
       "      <td>1.419593</td>\n",
       "      <td>1.420044</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.732064</td>\n",
       "      <td>1.734607</td>\n",
       "      <td>2.000047</td>\n",
       "      <td>2.000052</td>\n",
       "      <td>2.000136</td>\n",
       "      <td>2.000570</td>\n",
       "      <td>2.001120</td>\n",
       "      <td>2.001261</td>\n",
       "      <td>2.001281</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.033879</td>\n",
       "      <td>0.093164</td>\n",
       "      <td>0.112085</td>\n",
       "      <td>1.001799</td>\n",
       "      <td>1.002741</td>\n",
       "      <td>1.006064</td>\n",
       "      <td>1.016809</td>\n",
       "      <td>1.414255</td>\n",
       "      <td>1.414407</td>\n",
       "    </tr>\n",
       "    <tr>\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",
       "      <td>15233</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.414261</td>\n",
       "      <td>1.414820</td>\n",
       "      <td>1.414943</td>\n",
       "      <td>1.414990</td>\n",
       "      <td>1.416735</td>\n",
       "      <td>1.732052</td>\n",
       "      <td>1.732059</td>\n",
       "      <td>1.732070</td>\n",
       "      <td>1.732074</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15234</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.000004</td>\n",
       "      <td>1.001782</td>\n",
       "      <td>1.022514</td>\n",
       "      <td>1.414214</td>\n",
       "      <td>1.418791</td>\n",
       "      <td>1.418971</td>\n",
       "      <td>1.419427</td>\n",
       "      <td>1.419665</td>\n",
       "      <td>1.420146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15235</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.019463</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000004</td>\n",
       "      <td>1.000006</td>\n",
       "      <td>1.000027</td>\n",
       "      <td>1.000047</td>\n",
       "      <td>1.000092</td>\n",
       "      <td>1.000099</td>\n",
       "      <td>1.000103</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15236</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.000040</td>\n",
       "      <td>1.001170</td>\n",
       "      <td>1.002171</td>\n",
       "      <td>1.414255</td>\n",
       "      <td>1.414536</td>\n",
       "      <td>1.415354</td>\n",
       "      <td>1.415994</td>\n",
       "      <td>1.416181</td>\n",
       "      <td>1.416850</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15237</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.231374</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000212</td>\n",
       "      <td>1.000306</td>\n",
       "      <td>1.000622</td>\n",
       "      <td>1.414214</td>\n",
       "      <td>1.414217</td>\n",
       "      <td>1.414225</td>\n",
       "      <td>1.414231</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>15238 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         0         1         2         3         4         5         6  \\\n",
       "0      0.0  1.100676  1.108194  1.144710  1.478284  1.479574  1.481531   \n",
       "1      0.0  1.416066  1.418109  1.423655  1.426490  1.437977  1.481053   \n",
       "2      0.0  1.023500  1.415059  1.416404  1.416719  1.417555  1.418584   \n",
       "3      0.0  1.732064  1.734607  2.000047  2.000052  2.000136  2.000570   \n",
       "4      0.0  0.033879  0.093164  0.112085  1.001799  1.002741  1.006064   \n",
       "...    ...       ...       ...       ...       ...       ...       ...   \n",
       "15233  0.0  1.414261  1.414820  1.414943  1.414990  1.416735  1.732052   \n",
       "15234  0.0  1.000004  1.001782  1.022514  1.414214  1.418791  1.418971   \n",
       "15235  0.0  0.019463  1.000000  1.000004  1.000006  1.000027  1.000047   \n",
       "15236  0.0  1.000040  1.001170  1.002171  1.414255  1.414536  1.415354   \n",
       "15237  0.0  0.231374  1.000000  1.000212  1.000306  1.000622  1.414214   \n",
       "\n",
       "              7         8         9  \n",
       "0      1.482657  1.487760  1.488879  \n",
       "1      1.568456  1.733397  1.733798  \n",
       "2      1.418853  1.419593  1.420044  \n",
       "3      2.001120  2.001261  2.001281  \n",
       "4      1.016809  1.414255  1.414407  \n",
       "...         ...       ...       ...  \n",
       "15233  1.732059  1.732070  1.732074  \n",
       "15234  1.419427  1.419665  1.420146  \n",
       "15235  1.000092  1.000099  1.000103  \n",
       "15236  1.415994  1.416181  1.416850  \n",
       "15237  1.414217  1.414225  1.414231  \n",
       "\n",
       "[15238 rows x 10 columns]"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#printing distance matrixas dataframe for better visualization\n",
    "pd.DataFrame(distances1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Anime Recommendation for: Cowboy Bebop / Cowboy Bebop\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>Recommended Anime</th>\n",
       "      <th>English Title</th>\n",
       "      <th>Type</th>\n",
       "      <th>Rating</th>\n",
       "      <th>Genre</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>Cowboy Bebop</td>\n",
       "      <td>Cowboy Bebop</td>\n",
       "      <td>TV</td>\n",
       "      <td>8.81</td>\n",
       "      <td>Action, Adventure, Comedy, Drama, Sci-Fi, Space</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>Ginga Tetsudou Monogatari</td>\n",
       "      <td>The Galaxy Railways</td>\n",
       "      <td>TV</td>\n",
       "      <td>7.25</td>\n",
       "      <td>Action, Adventure, Space, Drama, Sci-Fi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>Ginga Tetsudou Monogatari: Eien e no Bunkiten</td>\n",
       "      <td>NaN</td>\n",
       "      <td>TV</td>\n",
       "      <td>6.98</td>\n",
       "      <td>Action, Adventure, Space, Drama, Sci-Fi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>Seihou Tenshi Angel Links</td>\n",
       "      <td>Angel Links</td>\n",
       "      <td>TV</td>\n",
       "      <td>6.02</td>\n",
       "      <td>Action, Adventure, Space, Comedy, Romance, Dra...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>Uchuu Patrol Luluco</td>\n",
       "      <td>Space Patrol Luluco</td>\n",
       "      <td>TV</td>\n",
       "      <td>7.68</td>\n",
       "      <td>Action, Adventure, Comedy, Space</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>Uchuu Senkan Yamato</td>\n",
       "      <td>Star Blazers: The Quest for Iscandar</td>\n",
       "      <td>TV</td>\n",
       "      <td>7.67</td>\n",
       "      <td>Action, Adventure, Drama, Military, Sci-Fi, Space</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>Uchuu Senkan Yamato 2</td>\n",
       "      <td>Star Blazers: The Comet Empire</td>\n",
       "      <td>TV</td>\n",
       "      <td>7.55</td>\n",
       "      <td>Action, Military, Sci-Fi, Adventure, Space, Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>Cowboy Bebop: Yose Atsume Blues</td>\n",
       "      <td>Cowboy Bebop Session XX</td>\n",
       "      <td>Special</td>\n",
       "      <td>7.49</td>\n",
       "      <td>Action, Adventure, Comedy, Drama, Sci-Fi, Space</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>Uchuu Senkan Yamato III</td>\n",
       "      <td>Star Blazers: The Bolar Wars</td>\n",
       "      <td>TV</td>\n",
       "      <td>7.21</td>\n",
       "      <td>Action, Military, Sci-Fi, Adventure, Space, Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>Uchuu Densetsu Ulysses 31</td>\n",
       "      <td>Ulysses 31</td>\n",
       "      <td>TV</td>\n",
       "      <td>7.15</td>\n",
       "      <td>Action, Adventure, Sci-Fi, Space</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                               Recommended Anime  \\\n",
       "0                                   Cowboy Bebop   \n",
       "1                      Ginga Tetsudou Monogatari   \n",
       "2  Ginga Tetsudou Monogatari: Eien e no Bunkiten   \n",
       "3                      Seihou Tenshi Angel Links   \n",
       "4                            Uchuu Patrol Luluco   \n",
       "5                            Uchuu Senkan Yamato   \n",
       "6                          Uchuu Senkan Yamato 2   \n",
       "7                Cowboy Bebop: Yose Atsume Blues   \n",
       "8                        Uchuu Senkan Yamato III   \n",
       "9                      Uchuu Densetsu Ulysses 31   \n",
       "\n",
       "                          English Title     Type  Rating  \\\n",
       "0                          Cowboy Bebop       TV    8.81   \n",
       "1                   The Galaxy Railways       TV    7.25   \n",
       "2                                   NaN       TV    6.98   \n",
       "3                           Angel Links       TV    6.02   \n",
       "4                   Space Patrol Luluco       TV    7.68   \n",
       "5  Star Blazers: The Quest for Iscandar       TV    7.67   \n",
       "6        Star Blazers: The Comet Empire       TV    7.55   \n",
       "7               Cowboy Bebop Session XX  Special    7.49   \n",
       "8          Star Blazers: The Bolar Wars       TV    7.21   \n",
       "9                            Ulysses 31       TV    7.15   \n",
       "\n",
       "                                               Genre  \n",
       "0    Action, Adventure, Comedy, Drama, Sci-Fi, Space  \n",
       "1            Action, Adventure, Space, Drama, Sci-Fi  \n",
       "2            Action, Adventure, Space, Drama, Sci-Fi  \n",
       "3  Action, Adventure, Space, Comedy, Romance, Dra...  \n",
       "4                   Action, Adventure, Comedy, Space  \n",
       "5  Action, Adventure, Drama, Military, Sci-Fi, Space  \n",
       "6  Action, Military, Sci-Fi, Adventure, Space, Drama  \n",
       "7    Action, Adventure, Comedy, Drama, Sci-Fi, Space  \n",
       "8  Action, Military, Sci-Fi, Adventure, Space, Drama  \n",
       "9                   Action, Adventure, Sci-Fi, Space  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "get_recommendation_Brute('Cowboy Bebop')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## BallTree Algorithm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Wall time: 19.1 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "nbrs2 = NearestNeighbors(n_neighbors=10, algorithm='ball_tree').fit(anime)\n",
    "distances2, indices2 = nbrs2.kneighbors(anime)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [],
   "source": [
    "#funtion to get recommendation from BallTree\n",
    "def get_recommendation_BallTree(name):\n",
    "    rec_id=df.index[df['name']==name].tolist()[0]\n",
    "    recommend = pd.DataFrame(columns = ['Recommended Anime', 'English Title', 'Type', 'Rating', 'Genre'])\n",
    "    for id in indices2[rec_id][:]:\n",
    "        new_row = {'Recommended Anime':df.iloc[id]['name'], 'English Title':df.iloc[id]['title_english'], 'Type':df.iloc[id]['type'], 'Rating':np.round(df.iloc[id]['rating'],2), 'Genre':df.iloc[id]['genre'] }\n",
    "        recommend = recommend.append(new_row, ignore_index=True)\n",
    "    print(\"Anime Recommendation for:\", name, '/',df.iloc[rec_id]['title_english'])\n",
    "    display(recommend)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.        , 1.10067573, 1.1081936 , ..., 1.48265702, 1.48775968,\n",
       "        1.4888789 ],\n",
       "       [0.        , 1.41606606, 1.41810946, ..., 1.56845625, 1.73339718,\n",
       "        1.73379833],\n",
       "       [0.        , 1.02350008, 1.41505862, ..., 1.4188526 , 1.41959297,\n",
       "        1.42004394],\n",
       "       ...,\n",
       "       [0.        , 0.01946333, 1.00000025, ..., 1.00009246, 1.00009917,\n",
       "        1.00010261],\n",
       "       [0.        , 1.0000399 , 1.00117049, ..., 1.41599442, 1.41618116,\n",
       "        1.41685002],\n",
       "       [0.        , 0.23137357, 1.00000025, ..., 1.41421672, 1.41422498,\n",
       "        1.41423112]])"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "distances2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.100676</td>\n",
       "      <td>1.108194</td>\n",
       "      <td>1.144710</td>\n",
       "      <td>1.478284</td>\n",
       "      <td>1.479574</td>\n",
       "      <td>1.481531</td>\n",
       "      <td>1.482657</td>\n",
       "      <td>1.487760</td>\n",
       "      <td>1.488879</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.416066</td>\n",
       "      <td>1.418109</td>\n",
       "      <td>1.423655</td>\n",
       "      <td>1.426490</td>\n",
       "      <td>1.437977</td>\n",
       "      <td>1.481053</td>\n",
       "      <td>1.568456</td>\n",
       "      <td>1.733397</td>\n",
       "      <td>1.733798</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.023500</td>\n",
       "      <td>1.415059</td>\n",
       "      <td>1.416404</td>\n",
       "      <td>1.416719</td>\n",
       "      <td>1.417555</td>\n",
       "      <td>1.418584</td>\n",
       "      <td>1.418853</td>\n",
       "      <td>1.419593</td>\n",
       "      <td>1.420044</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.732064</td>\n",
       "      <td>1.734607</td>\n",
       "      <td>2.000047</td>\n",
       "      <td>2.000052</td>\n",
       "      <td>2.000136</td>\n",
       "      <td>2.000570</td>\n",
       "      <td>2.001120</td>\n",
       "      <td>2.001261</td>\n",
       "      <td>2.001281</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.033879</td>\n",
       "      <td>0.093164</td>\n",
       "      <td>0.112085</td>\n",
       "      <td>1.001799</td>\n",
       "      <td>1.002741</td>\n",
       "      <td>1.006064</td>\n",
       "      <td>1.016809</td>\n",
       "      <td>1.414255</td>\n",
       "      <td>1.414407</td>\n",
       "    </tr>\n",
       "    <tr>\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",
       "      <td>15233</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.414261</td>\n",
       "      <td>1.414820</td>\n",
       "      <td>1.414943</td>\n",
       "      <td>1.414990</td>\n",
       "      <td>1.416735</td>\n",
       "      <td>1.732052</td>\n",
       "      <td>1.732059</td>\n",
       "      <td>1.732070</td>\n",
       "      <td>1.732074</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15234</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.000004</td>\n",
       "      <td>1.001782</td>\n",
       "      <td>1.022514</td>\n",
       "      <td>1.414214</td>\n",
       "      <td>1.418791</td>\n",
       "      <td>1.418971</td>\n",
       "      <td>1.419427</td>\n",
       "      <td>1.419665</td>\n",
       "      <td>1.420146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15235</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.019463</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000004</td>\n",
       "      <td>1.000006</td>\n",
       "      <td>1.000027</td>\n",
       "      <td>1.000047</td>\n",
       "      <td>1.000092</td>\n",
       "      <td>1.000099</td>\n",
       "      <td>1.000103</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15236</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.000040</td>\n",
       "      <td>1.001170</td>\n",
       "      <td>1.002171</td>\n",
       "      <td>1.414255</td>\n",
       "      <td>1.414536</td>\n",
       "      <td>1.415354</td>\n",
       "      <td>1.415994</td>\n",
       "      <td>1.416181</td>\n",
       "      <td>1.416850</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15237</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.231374</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000212</td>\n",
       "      <td>1.000306</td>\n",
       "      <td>1.000622</td>\n",
       "      <td>1.414214</td>\n",
       "      <td>1.414217</td>\n",
       "      <td>1.414225</td>\n",
       "      <td>1.414231</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>15238 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         0         1         2         3         4         5         6  \\\n",
       "0      0.0  1.100676  1.108194  1.144710  1.478284  1.479574  1.481531   \n",
       "1      0.0  1.416066  1.418109  1.423655  1.426490  1.437977  1.481053   \n",
       "2      0.0  1.023500  1.415059  1.416404  1.416719  1.417555  1.418584   \n",
       "3      0.0  1.732064  1.734607  2.000047  2.000052  2.000136  2.000570   \n",
       "4      0.0  0.033879  0.093164  0.112085  1.001799  1.002741  1.006064   \n",
       "...    ...       ...       ...       ...       ...       ...       ...   \n",
       "15233  0.0  1.414261  1.414820  1.414943  1.414990  1.416735  1.732052   \n",
       "15234  0.0  1.000004  1.001782  1.022514  1.414214  1.418791  1.418971   \n",
       "15235  0.0  0.019463  1.000000  1.000004  1.000006  1.000027  1.000047   \n",
       "15236  0.0  1.000040  1.001170  1.002171  1.414255  1.414536  1.415354   \n",
       "15237  0.0  0.231374  1.000000  1.000212  1.000306  1.000622  1.414214   \n",
       "\n",
       "              7         8         9  \n",
       "0      1.482657  1.487760  1.488879  \n",
       "1      1.568456  1.733397  1.733798  \n",
       "2      1.418853  1.419593  1.420044  \n",
       "3      2.001120  2.001261  2.001281  \n",
       "4      1.016809  1.414255  1.414407  \n",
       "...         ...       ...       ...  \n",
       "15233  1.732059  1.732070  1.732074  \n",
       "15234  1.419427  1.419665  1.420146  \n",
       "15235  1.000092  1.000099  1.000103  \n",
       "15236  1.415994  1.416181  1.416850  \n",
       "15237  1.414217  1.414225  1.414231  \n",
       "\n",
       "[15238 rows x 10 columns]"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#printing distance matrixas dataframe for better visualization\n",
    "pd.DataFrame(distances1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Anime Recommendation for: Cowboy Bebop / Cowboy Bebop\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>Recommended Anime</th>\n",
       "      <th>English Title</th>\n",
       "      <th>Type</th>\n",
       "      <th>Rating</th>\n",
       "      <th>Genre</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>Cowboy Bebop</td>\n",
       "      <td>Cowboy Bebop</td>\n",
       "      <td>TV</td>\n",
       "      <td>8.81</td>\n",
       "      <td>Action, Adventure, Comedy, Drama, Sci-Fi, Space</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>Ginga Tetsudou Monogatari</td>\n",
       "      <td>The Galaxy Railways</td>\n",
       "      <td>TV</td>\n",
       "      <td>7.25</td>\n",
       "      <td>Action, Adventure, Space, Drama, Sci-Fi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>Ginga Tetsudou Monogatari: Eien e no Bunkiten</td>\n",
       "      <td>NaN</td>\n",
       "      <td>TV</td>\n",
       "      <td>6.98</td>\n",
       "      <td>Action, Adventure, Space, Drama, Sci-Fi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>Seihou Tenshi Angel Links</td>\n",
       "      <td>Angel Links</td>\n",
       "      <td>TV</td>\n",
       "      <td>6.02</td>\n",
       "      <td>Action, Adventure, Space, Comedy, Romance, Dra...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>Uchuu Patrol Luluco</td>\n",
       "      <td>Space Patrol Luluco</td>\n",
       "      <td>TV</td>\n",
       "      <td>7.68</td>\n",
       "      <td>Action, Adventure, Comedy, Space</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>Uchuu Senkan Yamato</td>\n",
       "      <td>Star Blazers: The Quest for Iscandar</td>\n",
       "      <td>TV</td>\n",
       "      <td>7.67</td>\n",
       "      <td>Action, Adventure, Drama, Military, Sci-Fi, Space</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>Uchuu Senkan Yamato 2</td>\n",
       "      <td>Star Blazers: The Comet Empire</td>\n",
       "      <td>TV</td>\n",
       "      <td>7.55</td>\n",
       "      <td>Action, Military, Sci-Fi, Adventure, Space, Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>Cowboy Bebop: Yose Atsume Blues</td>\n",
       "      <td>Cowboy Bebop Session XX</td>\n",
       "      <td>Special</td>\n",
       "      <td>7.49</td>\n",
       "      <td>Action, Adventure, Comedy, Drama, Sci-Fi, Space</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>Uchuu Senkan Yamato III</td>\n",
       "      <td>Star Blazers: The Bolar Wars</td>\n",
       "      <td>TV</td>\n",
       "      <td>7.21</td>\n",
       "      <td>Action, Military, Sci-Fi, Adventure, Space, Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>Uchuu Densetsu Ulysses 31</td>\n",
       "      <td>Ulysses 31</td>\n",
       "      <td>TV</td>\n",
       "      <td>7.15</td>\n",
       "      <td>Action, Adventure, Sci-Fi, Space</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                               Recommended Anime  \\\n",
       "0                                   Cowboy Bebop   \n",
       "1                      Ginga Tetsudou Monogatari   \n",
       "2  Ginga Tetsudou Monogatari: Eien e no Bunkiten   \n",
       "3                      Seihou Tenshi Angel Links   \n",
       "4                            Uchuu Patrol Luluco   \n",
       "5                            Uchuu Senkan Yamato   \n",
       "6                          Uchuu Senkan Yamato 2   \n",
       "7                Cowboy Bebop: Yose Atsume Blues   \n",
       "8                        Uchuu Senkan Yamato III   \n",
       "9                      Uchuu Densetsu Ulysses 31   \n",
       "\n",
       "                          English Title     Type  Rating  \\\n",
       "0                          Cowboy Bebop       TV    8.81   \n",
       "1                   The Galaxy Railways       TV    7.25   \n",
       "2                                   NaN       TV    6.98   \n",
       "3                           Angel Links       TV    6.02   \n",
       "4                   Space Patrol Luluco       TV    7.68   \n",
       "5  Star Blazers: The Quest for Iscandar       TV    7.67   \n",
       "6        Star Blazers: The Comet Empire       TV    7.55   \n",
       "7               Cowboy Bebop Session XX  Special    7.49   \n",
       "8          Star Blazers: The Bolar Wars       TV    7.21   \n",
       "9                            Ulysses 31       TV    7.15   \n",
       "\n",
       "                                               Genre  \n",
       "0    Action, Adventure, Comedy, Drama, Sci-Fi, Space  \n",
       "1            Action, Adventure, Space, Drama, Sci-Fi  \n",
       "2            Action, Adventure, Space, Drama, Sci-Fi  \n",
       "3  Action, Adventure, Space, Comedy, Romance, Dra...  \n",
       "4                   Action, Adventure, Comedy, Space  \n",
       "5  Action, Adventure, Drama, Military, Sci-Fi, Space  \n",
       "6  Action, Military, Sci-Fi, Adventure, Space, Drama  \n",
       "7    Action, Adventure, Comedy, Drama, Sci-Fi, Space  \n",
       "8  Action, Military, Sci-Fi, Adventure, Space, Drama  \n",
       "9                   Action, Adventure, Sci-Fi, Space  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "get_recommendation_BallTree('Cowboy Bebop')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## KDTree Algorithm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Wall time: 14.6 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "nbrs3 = NearestNeighbors(n_neighbors=10, algorithm='kd_tree').fit(anime)\n",
    "distances3, indices3 = nbrs3.kneighbors(anime)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "#funtion to get recommendation from KDTree\n",
    "def get_recommendation_KDTree(name):\n",
    "    rec_id=df.index[df['name']==name].tolist()[0]\n",
    "    recommend = pd.DataFrame(columns = ['Recommended Anime', 'English Title', 'Type', 'Rating', 'Genre'])\n",
    "    for id in indices3[rec_id][:]:\n",
    "        new_row = {'Recommended Anime':df.iloc[id]['name'], 'English Title':df.iloc[id]['title_english'], 'Type':df.iloc[id]['type'], 'Rating':np.round(df.iloc[id]['rating'],2), 'Genre':df.iloc[id]['genre'] }\n",
    "        recommend = recommend.append(new_row, ignore_index=True)\n",
    "    print(\"Anime Recommendation for:\", name, '/',df.iloc[rec_id]['title_english'])\n",
    "    display(recommend)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.        , 1.10067573, 1.1081936 , ..., 1.48265702, 1.48775968,\n",
       "        1.4888789 ],\n",
       "       [0.        , 1.41606606, 1.41810946, ..., 1.56845625, 1.73339718,\n",
       "        1.73379833],\n",
       "       [0.        , 1.02350008, 1.41505862, ..., 1.4188526 , 1.41959297,\n",
       "        1.42004394],\n",
       "       ...,\n",
       "       [0.        , 0.01946333, 1.00000025, ..., 1.00009246, 1.00009917,\n",
       "        1.00010261],\n",
       "       [0.        , 1.0000399 , 1.00117049, ..., 1.41599442, 1.41618116,\n",
       "        1.41685002],\n",
       "       [0.        , 0.23137357, 1.00000025, ..., 1.41421672, 1.41422498,\n",
       "        1.41423112]])"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "distances3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.100676</td>\n",
       "      <td>1.108194</td>\n",
       "      <td>1.144710</td>\n",
       "      <td>1.478284</td>\n",
       "      <td>1.479574</td>\n",
       "      <td>1.481531</td>\n",
       "      <td>1.482657</td>\n",
       "      <td>1.487760</td>\n",
       "      <td>1.488879</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.416066</td>\n",
       "      <td>1.418109</td>\n",
       "      <td>1.423655</td>\n",
       "      <td>1.426490</td>\n",
       "      <td>1.437977</td>\n",
       "      <td>1.481053</td>\n",
       "      <td>1.568456</td>\n",
       "      <td>1.733397</td>\n",
       "      <td>1.733798</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.023500</td>\n",
       "      <td>1.415059</td>\n",
       "      <td>1.416404</td>\n",
       "      <td>1.416719</td>\n",
       "      <td>1.417555</td>\n",
       "      <td>1.418584</td>\n",
       "      <td>1.418853</td>\n",
       "      <td>1.419593</td>\n",
       "      <td>1.420044</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.732064</td>\n",
       "      <td>1.734607</td>\n",
       "      <td>2.000047</td>\n",
       "      <td>2.000052</td>\n",
       "      <td>2.000136</td>\n",
       "      <td>2.000570</td>\n",
       "      <td>2.001120</td>\n",
       "      <td>2.001261</td>\n",
       "      <td>2.001281</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.033879</td>\n",
       "      <td>0.093164</td>\n",
       "      <td>0.112085</td>\n",
       "      <td>1.001799</td>\n",
       "      <td>1.002741</td>\n",
       "      <td>1.006064</td>\n",
       "      <td>1.016809</td>\n",
       "      <td>1.414255</td>\n",
       "      <td>1.414407</td>\n",
       "    </tr>\n",
       "    <tr>\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",
       "      <td>15233</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.414261</td>\n",
       "      <td>1.414820</td>\n",
       "      <td>1.414943</td>\n",
       "      <td>1.414990</td>\n",
       "      <td>1.416735</td>\n",
       "      <td>1.732052</td>\n",
       "      <td>1.732059</td>\n",
       "      <td>1.732070</td>\n",
       "      <td>1.732074</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15234</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.000004</td>\n",
       "      <td>1.001782</td>\n",
       "      <td>1.022514</td>\n",
       "      <td>1.414214</td>\n",
       "      <td>1.418791</td>\n",
       "      <td>1.418971</td>\n",
       "      <td>1.419427</td>\n",
       "      <td>1.419665</td>\n",
       "      <td>1.420146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15235</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.019463</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000004</td>\n",
       "      <td>1.000006</td>\n",
       "      <td>1.000027</td>\n",
       "      <td>1.000047</td>\n",
       "      <td>1.000092</td>\n",
       "      <td>1.000099</td>\n",
       "      <td>1.000103</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15236</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.000040</td>\n",
       "      <td>1.001170</td>\n",
       "      <td>1.002171</td>\n",
       "      <td>1.414255</td>\n",
       "      <td>1.414536</td>\n",
       "      <td>1.415354</td>\n",
       "      <td>1.415994</td>\n",
       "      <td>1.416181</td>\n",
       "      <td>1.416850</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15237</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.231374</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000212</td>\n",
       "      <td>1.000306</td>\n",
       "      <td>1.000622</td>\n",
       "      <td>1.414214</td>\n",
       "      <td>1.414217</td>\n",
       "      <td>1.414225</td>\n",
       "      <td>1.414231</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>15238 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         0         1         2         3         4         5         6  \\\n",
       "0      0.0  1.100676  1.108194  1.144710  1.478284  1.479574  1.481531   \n",
       "1      0.0  1.416066  1.418109  1.423655  1.426490  1.437977  1.481053   \n",
       "2      0.0  1.023500  1.415059  1.416404  1.416719  1.417555  1.418584   \n",
       "3      0.0  1.732064  1.734607  2.000047  2.000052  2.000136  2.000570   \n",
       "4      0.0  0.033879  0.093164  0.112085  1.001799  1.002741  1.006064   \n",
       "...    ...       ...       ...       ...       ...       ...       ...   \n",
       "15233  0.0  1.414261  1.414820  1.414943  1.414990  1.416735  1.732052   \n",
       "15234  0.0  1.000004  1.001782  1.022514  1.414214  1.418791  1.418971   \n",
       "15235  0.0  0.019463  1.000000  1.000004  1.000006  1.000027  1.000047   \n",
       "15236  0.0  1.000040  1.001170  1.002171  1.414255  1.414536  1.415354   \n",
       "15237  0.0  0.231374  1.000000  1.000212  1.000306  1.000622  1.414214   \n",
       "\n",
       "              7         8         9  \n",
       "0      1.482657  1.487760  1.488879  \n",
       "1      1.568456  1.733397  1.733798  \n",
       "2      1.418853  1.419593  1.420044  \n",
       "3      2.001120  2.001261  2.001281  \n",
       "4      1.016809  1.414255  1.414407  \n",
       "...         ...       ...       ...  \n",
       "15233  1.732059  1.732070  1.732074  \n",
       "15234  1.419427  1.419665  1.420146  \n",
       "15235  1.000092  1.000099  1.000103  \n",
       "15236  1.415994  1.416181  1.416850  \n",
       "15237  1.414217  1.414225  1.414231  \n",
       "\n",
       "[15238 rows x 10 columns]"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#printing distance matrixas dataframe for better visualization\n",
    "pd.DataFrame(distances1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Anime Recommendation for: Cowboy Bebop / Cowboy Bebop\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>Recommended Anime</th>\n",
       "      <th>English Title</th>\n",
       "      <th>Type</th>\n",
       "      <th>Rating</th>\n",
       "      <th>Genre</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>Cowboy Bebop</td>\n",
       "      <td>Cowboy Bebop</td>\n",
       "      <td>TV</td>\n",
       "      <td>8.81</td>\n",
       "      <td>Action, Adventure, Comedy, Drama, Sci-Fi, Space</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>Ginga Tetsudou Monogatari</td>\n",
       "      <td>The Galaxy Railways</td>\n",
       "      <td>TV</td>\n",
       "      <td>7.25</td>\n",
       "      <td>Action, Adventure, Space, Drama, Sci-Fi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>Ginga Tetsudou Monogatari: Eien e no Bunkiten</td>\n",
       "      <td>NaN</td>\n",
       "      <td>TV</td>\n",
       "      <td>6.98</td>\n",
       "      <td>Action, Adventure, Space, Drama, Sci-Fi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>Seihou Tenshi Angel Links</td>\n",
       "      <td>Angel Links</td>\n",
       "      <td>TV</td>\n",
       "      <td>6.02</td>\n",
       "      <td>Action, Adventure, Space, Comedy, Romance, Dra...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>Uchuu Patrol Luluco</td>\n",
       "      <td>Space Patrol Luluco</td>\n",
       "      <td>TV</td>\n",
       "      <td>7.68</td>\n",
       "      <td>Action, Adventure, Comedy, Space</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>Uchuu Senkan Yamato</td>\n",
       "      <td>Star Blazers: The Quest for Iscandar</td>\n",
       "      <td>TV</td>\n",
       "      <td>7.67</td>\n",
       "      <td>Action, Adventure, Drama, Military, Sci-Fi, Space</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>Uchuu Senkan Yamato 2</td>\n",
       "      <td>Star Blazers: The Comet Empire</td>\n",
       "      <td>TV</td>\n",
       "      <td>7.55</td>\n",
       "      <td>Action, Military, Sci-Fi, Adventure, Space, Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>Cowboy Bebop: Yose Atsume Blues</td>\n",
       "      <td>Cowboy Bebop Session XX</td>\n",
       "      <td>Special</td>\n",
       "      <td>7.49</td>\n",
       "      <td>Action, Adventure, Comedy, Drama, Sci-Fi, Space</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>Uchuu Senkan Yamato III</td>\n",
       "      <td>Star Blazers: The Bolar Wars</td>\n",
       "      <td>TV</td>\n",
       "      <td>7.21</td>\n",
       "      <td>Action, Military, Sci-Fi, Adventure, Space, Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>Uchuu Densetsu Ulysses 31</td>\n",
       "      <td>Ulysses 31</td>\n",
       "      <td>TV</td>\n",
       "      <td>7.15</td>\n",
       "      <td>Action, Adventure, Sci-Fi, Space</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                               Recommended Anime  \\\n",
       "0                                   Cowboy Bebop   \n",
       "1                      Ginga Tetsudou Monogatari   \n",
       "2  Ginga Tetsudou Monogatari: Eien e no Bunkiten   \n",
       "3                      Seihou Tenshi Angel Links   \n",
       "4                            Uchuu Patrol Luluco   \n",
       "5                            Uchuu Senkan Yamato   \n",
       "6                          Uchuu Senkan Yamato 2   \n",
       "7                Cowboy Bebop: Yose Atsume Blues   \n",
       "8                        Uchuu Senkan Yamato III   \n",
       "9                      Uchuu Densetsu Ulysses 31   \n",
       "\n",
       "                          English Title     Type  Rating  \\\n",
       "0                          Cowboy Bebop       TV    8.81   \n",
       "1                   The Galaxy Railways       TV    7.25   \n",
       "2                                   NaN       TV    6.98   \n",
       "3                           Angel Links       TV    6.02   \n",
       "4                   Space Patrol Luluco       TV    7.68   \n",
       "5  Star Blazers: The Quest for Iscandar       TV    7.67   \n",
       "6        Star Blazers: The Comet Empire       TV    7.55   \n",
       "7               Cowboy Bebop Session XX  Special    7.49   \n",
       "8          Star Blazers: The Bolar Wars       TV    7.21   \n",
       "9                            Ulysses 31       TV    7.15   \n",
       "\n",
       "                                               Genre  \n",
       "0    Action, Adventure, Comedy, Drama, Sci-Fi, Space  \n",
       "1            Action, Adventure, Space, Drama, Sci-Fi  \n",
       "2            Action, Adventure, Space, Drama, Sci-Fi  \n",
       "3  Action, Adventure, Space, Comedy, Romance, Dra...  \n",
       "4                   Action, Adventure, Comedy, Space  \n",
       "5  Action, Adventure, Drama, Military, Sci-Fi, Space  \n",
       "6  Action, Military, Sci-Fi, Adventure, Space, Drama  \n",
       "7    Action, Adventure, Comedy, Drama, Sci-Fi, Space  \n",
       "8  Action, Military, Sci-Fi, Adventure, Space, Drama  \n",
       "9                   Action, Adventure, Sci-Fi, Space  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "get_recommendation_KDTree('Cowboy Bebop')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Unsupervised K-means"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Checking different score to determine the cluster quality for different K"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [],
   "source": [
    "SSE = [] #for inertia\n",
    "kmeSscore = [] #for Silhouette Coefficient\n",
    "kmeanCHscore = [] #for Calinski-Harabasz Index\n",
    "kmeanDBscore = [] #for Davies-Bouldin"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Current Progress: 100.0 %\n",
      "Wall time: 2min 41s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "for cluster in range(10,51,2):\n",
    "    clear_output(wait=True)\n",
    "    kmeans1 = KMeans(n_clusters = cluster, init='k-means++')\n",
    "    kmeans1.fit(anime)\n",
    "    SSE.append(kmeans1.inertia_)\n",
    "    kmeSscore.append(metrics.silhouette_score(anime,kmeans1.labels_, metric='euclidean'))\n",
    "    kmeanCHscore.append(metrics.calinski_harabasz_score(anime, kmeans1.labels_))\n",
    "    kmeanDBscore.append(davies_bouldin_score(anime, kmeans1.labels_))\n",
    "    print(\"Current Progress:\",np.round(cluster/50 *100, 2),\"%\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Inertia vs Cluster Graph"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Inertia')"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# converting the results into a dataframe and plotting them\n",
    "frame = pd.DataFrame({'Cluster':range(10,51,2), 'SSE':SSE})\n",
    "plt.figure(figsize=(20,10))\n",
    "plt.plot(frame['Cluster'], frame['SSE'], marker='o')\n",
    "plt.xlabel('Number of clusters')\n",
    "plt.ylabel('Inertia')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [],
   "source": [
    "#combining evaluation scores in a dataframe\n",
    "clu = range(10,51,2)\n",
    "dict1 = { 'Cluster Number': clu, 'Inertia': SSE, 'Silhouette Coefficient': kmeSscore, 'Calinski-Harabasz Index': kmeanCHscore, 'Davies-Bouldin': kmeanDBscore}\n",
    "score = pd.DataFrame(dict1)\n",
    "score.to_csv(r'D:\\NSU\\11th Semester\\CSE445 - AZK\\Project\\score.csv', index=False, mode='w')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Inertia"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Lowers values is better"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABJ4AAAJNCAYAAABwab9RAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOzdf6yeZ33f8c93NgG0liaAYWmczBHz2gakBuqFbEhbG9bgkK5JJ9CCpuJV2dxVYaJatWH6D+VHpqCpo0KjVGlxE7puIYJWsYhZlvFjVbcS4kAaCCmLG7zGTUbSJVAoW1jCd3+cy+PgHP/MuXyOz3m9pEfP81zPfd/nerj1IOet+0d1dwAAAABguf2llZ4AAAAAAGuT8AQAAADAFMITAAAAAFMITwAAAABMITwBAAAAMIXwBAAAAMAUG1d6AqfaC1/4wt6yZctKTwMAAABgzbjrrrv+rLs3HT6+7sLTli1bsm/fvpWeBgAAAMCaUVX/Y6lxp9oBAAAAMIXwBAAAAMAUwhMAAAAAUwhPAAAAAEwhPAEAAAAwhfAEAAAAwBTCEwAAAABTCE8AAAAATCE8AQAAADCF8AQAAADAFMITAAAAAFMITwAAAABMITwBAAAAMIXwBAAAAMAUwhMAAAAAUwhPAAAAAEwhPAEAAAAwhfAEAAAAwBTCEwAAAABTCE8AAAAATCE8AQAAADCF8AQAAADAFMITAAAAAFMITwAAAABMsXGlJ8DJ2bLr1mXd3oHrLl/W7QEAAAA44gkAAACAKYQnAAAAAKaYHp6qakNVfa6qPjren19Vd1TV/VX1oao6Y4w/e7zfPz7fsmgbbx3jX6qq1ywa3z7G9lfVrtnfBQAAAIDjdyqOeHpzkvsWvX93kvd099Ykjye5eoxfneTx7v5rSd4zlktVXZDkqiQvTbI9ya+OmLUhyfuSXJbkgiRvGMsCAAAAsApMDU9VtTnJ5Ul+Y7yvJJck+fBY5MYkV47XV4z3GZ+/eix/RZKbuvuJ7v5ykv1JLhqP/d39QHd/K8lNY1kAAAAAVoHZRzz9SpJ/meTb4/0Lkny1u58c7w8mOWe8PifJg0kyPv/aWP7/jx+2zpHGAQAAAFgFpoWnqvqJJI90912Lh5dYtI/x2YmOLzWXnVW1r6r2Pfroo0eZNQAAAADLZeYRT69K8pNVdSALp8FdkoUjoM6sqo1jmc1JHhqvDyY5N0nG59+X5LHF44etc6Txp+nu67t7W3dv27Rp0zP/ZgAAAAAc07Tw1N1v7e7N3b0lCxcH/0R3/8Mkn0zyurHYjiS3jNd7xvuMzz/R3T3Grxp3vTs/ydYkn0lyZ5Kt4y55Z4y/sWfW9wEAAADgxGw89iLL7i1JbqqqdyX5XJIPjPEPJPmtqtqfhSOdrkqS7r63qm5O8sUkTya5prufSpKqelOS25JsSLK7u+89pd8EAAAAgCM6JeGpuz+V5FPj9QNZuCPd4cv8nySvP8L61ya5donxvUn2LuNUAQAAAFgms+9qBwAAAMA6JTwBAAAAMIXwBAAAAMAUwhMAAAAAUwhPAAAAAEwhPAEAAAAwhfAEAAAAwBTCEwAAAABTbFzpCbC+bdl167Ju78B1ly/r9gAAAICT54gnAAAAAKYQngAAAACYQngCAAAAYArhCQAAAIAphCcAAAAAphCeAAAAAJhCeAIAAABgCuEJAAAAgCmEJwAAAACmEJ4AAAAAmEJ4AgAAAGAK4QkAAACAKTau9ARgLdiy69Zl3d6B6y5f1u0BAADASnDEEwAAAABTCE8AAAAATCE8AQAAADCF8AQAAADAFMITAAAAAFMITwAAAABMITwBAAAAMIXwBAAAAMAUwhMAAAAAUwhPAAAAAEwhPAEAAAAwxcaVngCwemzZdeuybu/AdZcv6/YAAAA4vTjiCQAAAIAphCcAAAAAphCeAAAAAJhCeAIAAABgCuEJAAAAgCmEJwAAAACmEJ4AAAAAmEJ4AgAAAGCKjSs9AYDltGXXrcu6vQPXXb6s2wMAAFhPHPEEAAAAwBTCEwAAAABTCE8AAAAATCE8AQAAADCF8AQAAADAFMITAAAAAFNsXOkJAPB0W3bduqzbO3Dd5cu6PQAAgOPhiCcAAAAAphCeAAAAAJhCeAIAAABgCuEJAAAAgCmEJwAAAACmmBaequo5VfWZqvrDqrq3qt4+xm+oqi9X1d3jceEYr6p6b1Xtr6p7quoVi7a1o6ruH48di8Z/pKo+P9Z5b1XVrO8DAAAAwInZOHHbTyS5pLu/UVXPSvL7VfWx8dm/6O4PH7b8ZUm2jscrk7w/ySur6vlJ3pZkW5JOcldV7enux8cyO5N8OsneJNuTfCwAAAAArLhpRzz1gm+Mt88ajz7KKlck+eBY79NJzqyqs5O8Jsnt3f3YiE23J9k+Pnted/9Bd3eSDya5ctb3AQAAAODETL3GU1VtqKq7kzyShXh0x/jo2nE63Xuq6tlj7JwkDy5a/eAYO9r4wSXGAQAAAFgFZp5ql+5+KsmFVXVmkt+tqpcleWuS/5nkjCTXJ3lLknckWer6TH0S409TVTuzcEpezjvvvBP8FgCcrC27bl3W7R247vJl3R4AADDXKbmrXXd/Ncmnkmzv7ofH6XRPJPnNJBeNxQ4mOXfRapuTPHSM8c1LjC/196/v7m3dvW3Tpk3L8I0AAAAAOJaZd7XbNI50SlU9N8nfTfJH49pMGXeguzLJF8Yqe5K8cdzd7uIkX+vuh5PcluTSqjqrqs5KcmmS28ZnX6+qi8e23pjkllnfBwAAAIATM/NUu7OT3FhVG7IQuG7u7o9W1SeqalMWTpW7O8k/HcvvTfLaJPuTfDPJzyRJdz9WVe9McudY7h3d/dh4/XNJbkjy3Czczc4d7QAAAABWiWnhqbvvSfLyJcYvOcLyneSaI3y2O8nuJcb3JXnZM5spAAAAADOckms8AQAAALD+CE8AAAAATDHzGk8AsCZs2XXrsm3rwHWXL9u2AABgtXPEEwAAAABTCE8AAAAATCE8AQAAADCF8AQAAADAFMITAAAAAFMITwAAAABMITwBAAAAMMXGlZ4AADDfll23Luv2Dlx3+bJuDwCAtckRTwAAAABMITwBAAAAMIXwBAAAAMAUwhMAAAAAUwhPAAAAAEwhPAEAAAAwxcaVngAAwPHasuvWZd3egesuX9btAQDw3RzxBAAAAMAUwhMAAAAAUwhPAAAAAEzhGk8AAKeYa1UBAOuFI54AAAAAmEJ4AgAAAGAK4QkAAACAKYQnAAAAAKYQngAAAACYwl3tAAA4Ye7MBwAcD0c8AQAAADCF8AQAAADAFMITAAAAAFMITwAAAABMITwBAAAAMIW72gEAsG65Ox8AzOWIJwAAAACmEJ4AAAAAmEJ4AgAAAGAK4QkAAACAKYQnAAAAAKYQngAAAACYQngCAAAAYIqNKz0BAADgmdmy69Zl3d6B6y5f1u0BsH454gkAAACAKYQnAAAAAKYQngAAAACYwjWeAACAVcG1qgDWHuEJAABgmYhnAN/NqXYAAAAATCE8AQAAADCF8AQAAADAFK7xBAAAwHdxrSpguTjiCQAAAIAphCcAAAAAphCeAAAAAJhCeAIAAABgimnhqaqeU1Wfqao/rKp7q+rtY/z8qrqjqu6vqg9V1Rlj/Nnj/f7x+ZZF23rrGP9SVb1m0fj2Mba/qnbN+i4AAAAAnLiZd7V7Iskl3f2NqnpWkt+vqo8l+edJ3tPdN1XVryW5Osn7x/Pj3f3XquqqJO9O8g+q6oIkVyV5aZLvT/Kfq+qvj7/xviQ/nuRgkjurak93f3HidwIAAOA04e58sPKmHfHUC74x3j5rPDrJJUk+PMZvTHLleH3FeJ/x+aurqsb4Td39RHd/Ocn+JBeNx/7ufqC7v5XkprEsAAAAAKvA1Gs8VdWGqro7ySNJbk/yx0m+2t1PjkUOJjlnvD4nyYNJMj7/WpIXLB4/bJ0jjQMAAACwCkwNT939VHdfmGRzFo5Q+qGlFhvPdYTPTnT8aapqZ1Xtq6p9jz766LEnDgAAAMAzNvMaT/9fd3+1qj6V5OIkZ1bVxnFU0+YkD43FDiY5N8nBqtqY5PuSPLZo/JDF6xxp/PC/f32S65Nk27ZtS8YpAAAAWI1cq4rT2cy72m2qqjPH6+cm+btJ7kvyySSvG4vtSHLLeL1nvM/4/BPd3WP8qnHXu/OTbE3ymSR3Jtk67pJ3RhYuQL5n1vcBAAAA4MTMPOLp7CQ3VtWGLASum7v7o1X1xSQ3VdW7knwuyQfG8h9I8ltVtT8LRzpdlSTdfW9V3Zzki0meTHJNdz+VJFX1piS3JdmQZHd33zvx+wAAAAAnwVFb69e08NTd9yR5+RLjD2Thek+Hj/+fJK8/wrauTXLtEuN7k+x9xpMFAAAAYNlNvbg4AAAAAOuX8AQAAADAFMITAAAAAFPMvLg4AAAAwJrhIuknzhFPAAAAAEwhPAEAAAAwhfAEAAAAwBTCEwAAAABTCE8AAAAATCE8AQAAADCF8AQAAADAFMITAAAAAFMITwAAAABMITwBAAAAMIXwBAAAAMAUwhMAAAAAUwhPAAAAAEwhPAEAAAAwhfAEAAAAwBTCEwAAAABTCE8AAAAATCE8AQAAADCF8AQAAADAFMITAAAAAFMITwAAAABMITwBAAAAMIXwBAAAAMAUwhMAAAAAUwhPAAAAAEwhPAEAAAAwhfAEAAAAwBTCEwAAAABTCE8AAAAATCE8AQAAADCF8AQAAADAFMITAAAAAFMITwAAAABMITwBAAAAMIXwBAAAAMAUwhMAAAAAUwhPAAAAAEwhPAEAAAAwhfAEAAAAwBTCEwAAAABTCE8AAAAATCE8AQAAADCF8AQAAADAFMITAAAAAFMITwAAAABMITwBAAAAMIXwBAAAAMAUwhMAAAAAUwhPAAAAAEwhPAEAAAAwhfAEAAAAwBTTwlNVnVtVn6yq+6rq3qp68xj/par606q6ezxeu2idt1bV/qr6UlW9ZtH49jG2v6p2LRo/v6ruqKr7q+pDVXXGrO8DAAAAwImZecTTk0l+obt/KMnFSa6pqgvGZ+/p7gvHY2+SjM+uSvLSJNuT/GpVbaiqDUnel+SyJBckecOi7bx7bGtrkseTXD3x+wAAAABwAqaFp+5+uLs/O15/Pcl9Sc45yipXJLmpu5/o7i8n2Z/kovHY390PdPe3ktyU5IqqqiSXJPnwWP/GJFfO+TYAAAAAnKhTco2nqtqS5OVJ7hhDb6qqe6pqd1WdNcbOSfLgotUOjrEjjb8gyVe7+8nDxgEAAABYBaaHp6r6niQfSfLz3f3nSd6f5CVJLkzycJJfPrToEqv3SYwvNYedVbWvqvY9+uijJ/gNAAAAADgZU8NTVT0rC9Hpt7v7d5Kku7/S3U9197eT/HoWTqVLFo5YOnfR6puTPHSU8T9LcmZVbTxs/Gm6+/ru3tbd2zZt2rQ8Xw4AAACAo5p5V7tK8oEk93X3v1k0fvaixX4qyRfG6z1JrqqqZ1fV+Um2JvlMkjuTbB13sDsjCxcg39PdneSTSV431t+R5JZZ3wcAAACAE7Px2IuctFcl+ekkn6+qu8fYL2bhrnQXZuG0uANJfjZJuvveqro5yRezcEe8a7r7qSSpqjcluS3JhiS7u/vesb23JLmpqt6V5HNZCF0AAAAArALTwlN3/36Wvg7T3qOsc22Sa5cY37vUet39QL5zqh4AAAAAq8gpuasdAAAAAOuP8AQAAADAFMITAAAAAFMITwAAAABMITwBAAAAMIXwBAAAAMAUwhMAAAAAUwhPAAAAAEwhPAEAAAAwhfAEAAAAwBTCEwAAAABTCE8AAAAATCE8AQAAADCF8AQAAADAFMITAAAAAFMITwAAAABMITwBAAAAMIXwBAAAAMAUwhMAAAAAUwhPAAAAAEwhPAEAAAAwhfAEAAAAwBTCEwAAAABTCE8AAAAATCE8AQAAADCF8AQAAADAFMITAAAAAFMITwAAAABMsfF4F6yqy5O8NMlzDo119ztmTAoAAACA099xHfFUVb+W5B8k+WdJKsnrk/zVifMCAAAA4DR3vKfa/a3ufmOSx7v77Un+ZpJz500LAAAAgNPd8Yan/z2ev1lV35/k/yY5f86UAAAAAFgLjvcaTx+tqjOT/Oskn03SSX5j2qwAAAAAOO0dV3jq7neOlx+pqo8meU53f23etAAAAAA43R01PFXVJd39iar6+0t8lu7+nXlTAwAAAOB0dqwjnv5Okk8k+XtLfNZJhCcAAAAAlnTU8NTdbxsv39HdX178WVW5uDgAAAAAR3S8d7X7yBJjH17OiQAAAACwthzrGk8/mOSlSb7vsOs8PS/Jc2ZODAAAAIDT27Gu8fQDSX4iyZn57us8fT3JP5k1KQAAAABOf8e6xtMtVfXRJG/p7n91iuYEAAAAwBpwzGs8dfdTSX78FMwFAAAAgDXkWKfaHfLfqurfJvlQkr84NNjdn50yKwAAAABOe8cbnv7WeH7HorFOcsnyTgcAAACAteK4wlN3/9jsiQAAAACwthzzGk9JUlUvrqoPVNXHxvsLqurquVMDAAAA4HR2XOEpyQ1Jbkvy/eP9f0/y8zMmBAAAAMDacLzh6YXdfXOSbydJdz+Z5KlpswIAAADgtHe84ekvquoFWbigeKrq4iRfmzYrAAAAAE57x3tXu3+eZE+Sl1TVf02yKcnrps0KAAAAgNPe8d7V7rNV9XeS/ECSSvKl7v6/U2cGAAAAwGnteI94SpKLkmwZ67yiqtLdH5wyKwAAAABOe8cVnqrqt5K8JMnd+c5FxTuJ8AQAAADAko73iKdtSS7o7p45GQAAAADWjuO9q90XkvyVmRMBAAAAYG053vD0wiRfrKrbqmrPocfRVqiqc6vqk1V1X1XdW1VvHuPPr6rbq+r+8XzWGK+qem9V7a+qe6rqFYu2tWMsf39V7Vg0/iNV9fmxznurqk78fwIAAAAAZjjeU+1+6SS2/WSSXxh3xPveJHdV1e1J/lGSj3f3dVW1K8muJG9JclmSrePxyiTvT/LKqnp+krdl4XS/HtvZ092Pj2V2Jvl0kr1Jtif52EnMFQAAAIBldlzhqbv/y4luuLsfTvLweP31qrovyTlJrkjyo2OxG5N8Kgvh6YokHxzXkfp0VZ1ZVWePZW/v7seSZMSr7VX1qSTP6+4/GOMfTHJlhCcAAACAVeGo4amqvp6Fo4ye9lGS7u7nHc8fqaotSV6e5I4kLx5RKt39cFW9aCx2TpIHF612cIwdbfzgEuMAAAAArAJHDU/d/b3P9A9U1fck+UiSn+/uPz/KZZiW+qBPYnypOezMwil5Oe+88441ZQAAAACWwfFeXPykVNWzshCdfru7f2cMf2WcQpfx/MgYP5jk3EWrb07y0DHGNy8x/jTdfX13b+vubZs2bXpmXwoAAACA4zItPI07zH0gyX3d/W8WfbQnyaE70+1Icsui8TeOu9tdnORr45S825JcWlVnjTvgXZrktvHZ16vq4vG33rhoWwAAAACssOO9q93JeFWSn07y+aq6e4z9YpLrktxcVVcn+ZMkrx+f7U3y2iT7k3wzyc8kSXc/VlXvTHLnWO4dhy40nuTnktyQ5LlZuKi4C4sDAAAArBLTwlN3/36Wvg5Tkrx6ieU7yTVH2NbuJLuXGN+X5GXPYJoAAAAATDL1Gk8AAAAArF/CEwAAAABTCE8AAAAATCE8AQAAADCF8AQAAADAFMITAAAAAFMITwAAAABMITwBAAAAMIXwBAAAAMAUwhMAAAAAUwhPAAAAAEwhPAEAAAAwhfAEAAAAwBTCEwAAAABTCE8AAAAATCE8AQAAADCF8AQAAADAFMITAAAAAFMITwAAAABMITwBAAAAMIXwBAAAAMAUwhMAAAAAUwhPAAAAAEwhPAEAAAAwhfAEAAAAwBTCEwAAAABTCE8AAAAATCE8AQAAADCF8AQAAADAFMITAAAAAFMITwAAAABMITwBAAAAMIXwBAAAAMAUwhMAAAAAUwhPAAAAAEwhPAEAAAAwhfAEAAAAwBTCEwAAAABTCE8AAAAATCE8AQAAADCF8AQAAADAFMITAAAAAFMITwAAAABMITwBAAAAMIXwBAAAAMAUwhMAAAAAUwhPAAAAAEwhPAEAAAAwhfAEAAAAwBTCEwAAAABTCE8AAAAATCE8AQAAADCF8AQAAADAFMITAAAAAFNMC09VtbuqHqmqLywa+6Wq+tOquns8Xrvos7dW1f6q+lJVvWbR+PYxtr+qdi0aP7+q7qiq+6vqQ1V1xqzvAgAAAMCJm3nE0w1Jti8x/p7uvnA89iZJVV2Q5KokLx3r/GpVbaiqDUnel+SyJBckecNYNknePba1NcnjSa6e+F0AAAAAOEHTwlN3/16Sx45z8SuS3NTdT3T3l5PsT3LReOzv7ge6+1tJbkpyRVVVkkuSfHisf2OSK5f1CwAAAADwjKzENZ7eVFX3jFPxzhpj5yR5cNEyB8fYkcZfkOSr3f3kYeMAAAAArBKnOjy9P8lLklyY5OEkvzzGa4ll+yTGl1RVO6tqX1Xte/TRR09sxgAAAACclFManrr7K939VHd/O8mvZ+FUumThiKVzFy26OclDRxn/syRnVtXGw8aP9Hev7+5t3b1t06ZNy/NlAAAAADiqUxqeqursRW9/KsmhO97tSXJVVT27qs5PsjXJZ5LcmWTruIPdGVm4APme7u4kn0zyurH+jiS3nIrvAAAAAMDx2XjsRU5OVf2HJD+a5IVVdTDJ25L8aFVdmIXT4g4k+dkk6e57q+rmJF9M8mSSa7r7qbGdNyW5LcmGJLu7+97xJ96S5KaqeleSzyX5wKzvAgAAAMCJmxaeuvsNSwwfMQ5197VJrl1ifG+SvUuMP5DvnKoHAAAAwCqzEne1AwAAAGAdEJ4AAAAAmEJ4AgAAAGAK4QkAAACAKYQnAAAAAKYQngAAAACYQngCAAAAYArhCQAAAIAphCcAAAAAphCeAAAAAJhCeAIAAABgCuEJAAAAgCmEJwAAAACmEJ4AAAAAmEJ4AgAAAGAK4QkAAACAKYQnAAAAAKYQngAAAACYQngCAAAAYArhCQAAAIAphCcAAAAAphCeAAAAAJhCeAIAAABgCuEJAAAAgCmEJwAAAACmEJ4AAAAAmEJ4AgAAAGAK4QkAAACAKYQnAAAAAKYQngAAAACYQngCAAAAYArhCQAAAIAphCcAAAAAphCeAAAAAJhCeAIAAABgCuEJAAAAgCmEJwAAAACmEJ4AAAAAmEJ4AgAAAGAK4QkAAACAKYQnAAAAAKYQngAAAACYQngCAAAAYArhCQAAAIAphCcAAAAAphCeAAAAAJhCeAIAAABgCuEJAAAAgCmEJwAAAACmEJ4AAAAAmEJ4AgAAAGAK4QkAAACAKYQnAAAAAKYQngAAAACYQngCAAAAYIpp4amqdlfVI1X1hUVjz6+q26vq/vF81hivqnpvVe2vqnuq6hWL1tkxlr+/qnYsGv+Rqvr8WOe9VVWzvgsAAAAAJ27mEU83JNl+2NiuJB/v7q1JPj7eJ8llSbaOx84k708WQlWStyV5ZZKLkrztUKway+xctN7hfwsAAACAFTQtPHX37yV57LDhK5LcOF7fmOTKReMf7AWfTnJmVZ2d5DVJbu/ux7r78SS3J9k+Pnted/9Bd3eSDy7aFgAAAACrwKm+xtOLu/vhJBnPLxrj5yR5cNFyB8fY0cYPLjEOAAAAwCqxWi4uvtT1mfokxpfeeNXOqtpXVfseffTRk5wiAAAAACfiVIenr4zT5DKeHxnjB5Ocu2i5zUkeOsb45iXGl9Td13f3tu7etmnTpmf8JQAAAAA4tlMdnvYkOXRnuh1Jblk0/sZxd7uLk3xtnIp3W5JLq+qscVHxS5PcNj77elVdPO5m98ZF2wIAAABgFdg4a8NV9R+S/GiSF1bVwSzcne66JDdX1dVJ/iTJ68fie5O8Nsn+JN9M8jNJ0t2PVdU7k9w5lntHdx+6YPnPZeHOec9N8rHxAAAAAGCVmBaeuvsNR/jo1Uss20muOcJ2difZvcT4viQveyZzBAAAAGCe1XJxcQAAAADWGOEJAAAAgCmEJwAAAACmEJ4AAAAAmEJ4AgAAAGAK4QkAAACAKYQnAAAAAKYQngAAAACYQngCAAAAYArhCQAAAIAphCcAAAAAphCeAAAAAJhCeAIAAABgCuEJAAAAgCmEJwAAAACmEJ4AAAAAmEJ4AgAAAGAK4QkAAACAKYQnAAAAAKYQngAAAACYQngCAAAAYArhCQAAAIAphCcAAAAAphCeAAAAAJhCeAIAAABgCuEJAAAAgCmEJwAAAACmEJ4AAAAAmEJ4AgAAAGAK4QkAAACAKYQnAAAAAKYQngAAAACYQngCAAAAYArhCQAAAIAphCcAAAAAphCeAAAAAJhCeAIAAABgCuEJAAAAgCmEJwAAAACmEJ4AAAAAmEJ4AgAAAGAK4QkAAACAKYQnAAAAAKYQngAAAACYQngCAAAAYArhCQAAAIAphCcAAAAAphCeAAAAAJhCeAIAAABgCuEJAAAAgCmEJwAAAACmEJ4AAAAAmEJ4AgAAAGAK4QkAAACAKYQnAAAAAKZYkfBUVQeq6vNVdXdV7Rtjz6+q26vq/vF81hivqnpvVe2vqnuq6hWLtrNjLH9/Ve1Yie8CAAAAwNJW8oinH+vuC7t723i/K8nHu3trko+P90lyWZKt47EzyfuThVCV5G1JXpnkoiRvOxSrAAAAAFh5q+lUuyuS3Dhe35jkykXjH+wFn05yZlWdneQ1SW7v7se6+/EktyfZfqonDQAAAMDSVio8dZL/VFV3VdXOMfbi7n44Scbzi8b4OUkeXLTuwTF2pHEAAAAAVoGNK/R3X9XdD1XVi5LcXlV/dJRla4mxPsr40zewELd2Jsl55513onMFAAAA4CSsyBFP3f3QeH4kye9m4RpNXxmn0GU8PzIWP5jk3EWrb07y0FHGl/p713f3tu7etmnTpuX8KgAAAAAcwSkPT1X1l6vqew+9TnJpki8k2ZPk0J3pdiS5Zbzek+SN4+52Fyf52jgV77Ykl1bVWeOi4peOMQAAAABWgZU41e7FSX63qmNOpv0AAAyASURBVA79/X/f3f+xqu5McnNVXZ3kT5K8fiy/N8lrk+xP8s0kP5Mk3f1YVb0zyZ1juXd092On7msAAAAAcDSnPDx19wNJfniJ8f+V5NVLjHeSa46wrd1Jdi/3HAEAAAB45lbqrnYAAAAArHHCEwAAAABTCE8AAAAATCE8AQAAADCF8AQAAADAFMITAAAAAFMITwAAAABMITwBAAAAMIXwBAAAAMAUwhMAAAAAUwhPAAAAAEwhPAEAAAAwhfAEAAAAwBTCEwAAAABTCE8AAAAATCE8AQAAADCF8AQAAADAFMITAAAAAFMITwAAAABMITwBAAAAMIXwBAAAAMAUwhMAAAAAUwhPAAAAAEwhPAEAAAAwhfAEAAAAwBTCEwAAAABTCE8AAAAATCE8AQAAADCF8AQAAADAFMITAAAAAFMITwAAAABMITwBAAAAMIXwBAAAAMAUwhMAAAAAUwhPAAAAAEwhPAEAAAAwhfAEAAAAwBTCEwAAAABTCE8AAAAATCE8AQAAADCF8AQAAADAFMITAAAAAFMITwAAAABMITwBAAAAMIXwBAAAAMAUwhMAAAAAUwhPAAAAAEwhPAEAAAAwhfAEAAAAwBTCEwAAAABTCE8AAAAATCE8AQAAADCF8AQAAADAFMITAAAAAFMITwAAAABMcdqHp6raXlVfqqr9VbVrpecDAAAAwILTOjxV1YYk70tyWZILkryhqi5Y2VkBAAAAkJzm4SnJRUn2d/cD3f2tJDcluWKF5wQAAABATv/wdE6SBxe9PzjGAAAAAFhh1d0rPYeTVlWvT/Ka7v7H4/1PJ7mou//ZYcvtTLJzvP2BJF86pRNdWS9M8mcrPQlOKft8fbG/1xf7e/2xz9cX+3t9sb/XH/t8fVmP+/uvdvemwwc3rsRMltHBJOcuer85yUOHL9Td1ye5/lRNajWpqn3dvW2l58GpY5+vL/b3+mJ/rz/2+fpif68v9vf6Y5+vL/b3d5zup9rdmWRrVZ1fVWckuSrJnhWeEwAAAAA5zY946u4nq+pNSW5LsiHJ7u6+d4WnBQAAAEBO8/CUJN29N8nelZ7HKrYuTzFc5+zz9cX+Xl/s7/XHPl9f7O/1xf5ef+zz9cX+Hk7ri4sDAAAAsHqd7td4AgAAAGCVEp7WkKraXVWPVNUXFo09v6pur6r7x/NZKzlHls8R9vcvVdWfVtXd4/HalZwjy6eqzq2qT1bVfVV1b1W9eYz7ja9RR9nnfudrUFU9p6o+U1V/OPb328f4+VV1x/iNf2jcTIXT3FH29w1V9eVFv+8LV3quLK+q2lBVn6uqj473fuNr2BL72298DauqA1X1+bFv940x/1aP8LTW3JBk+2Fju5J8vLu3Jvn4eM/acEOevr+T5D3dfeF4uP7Z2vFkkl/o7h9KcnGSa6rqgviNr2VH2ueJ3/la9ESSS7r7h5NcmGR7VV2c5N1Z2N9bkzye5OoVnCPL50j7O0n+xaLf990rN0UmeXOS+xa99xtf2w7f34nf+Fr3Y2Pfbhvv/Vs9wtOa0t2/l+Sxw4avSHLjeH1jkitP6aSY5gj7mzWqux/u7s+O11/Pwj9izonf+Jp1lH3OGtQLvjHePms8OsklST48xv3G14ij7G/WsKranOTyJL8x3lf8xtesw/c365Z/q0d4Wg9e3N0PJwv/EZPkRSs8H+Z7U1XdM07FW5eHcq51VbUlycuT3BG/8XXhsH2e+J2vSeOUjLuTPJLk9iR/nOSr3f3kWORgxMc14/D93d2Hft/Xjt/3e6rq2Ss4RZbfryT5l0m+Pd6/IH7ja9nh+/sQv/G1q5P8p6q6q6p2jjH/Vo/wBGvN+5O8JAuH7T+c5JdXdjost6r6niQfSfLz3f3nKz0f5ltin/udr1Hd/VR3X5hkc5KLkvzQUoud2lkxy+H7u6peluStSX4wyd9I8vwkb1nBKbKMquonkjzS3XctHl5iUb/xNeAI+zvxG1/rXtXdr0hyWRYukfC3V3pCq4XwtPZ9parOTpLx/MgKz4eJuvsr4x+y307y61n4DxfWiKp6VhYCxG939++MYb/xNWypfe53vvZ191eTfCoL1/Y6s6o2jo82J3lopebFHIv29/Zxim139xNJfjN+32vJq5L8ZFUdSHJTFk6x+5X4ja9VT9vfVfXv/MbXtu5+aDw/kuR3s7B//Vs9wtN6sCfJjvF6R5JbVnAuTHbo/9SGn0ryhSMt+//au/9QPcs6juPvj5tMmSMjlxkhE2kp/mhUVqjDVWNYShTOfrgywUjFCqWBuiIJAjfNhQVaibbIpukfIa6yUS6sRbY2zzpuWlJbJoIUkdXE1Tzf/rjvkzeH8+yc487jOM/eLzhwcd3Xz3NzwXO+57quRzNLew/EHcDjVbW288g1PqB6vXPX+WBKMj/J0W36SGApzb1em4DlbTHX+IDo8b6f6PxxEpp7QFzfA6KqrquqN1TVAuAjwENVtQLX+EDq8b4/5hofXEnmJpk3mgaW0bxfP6sDsycuopkiyd3AEuCYJE8D1wOrgXuTXAo8BVx48Eao6dTjfS9pv5a1gN3AZQdtgJpuZwEfB4bbO0EAVuEaH2S93vlHXecD6TjgO0lm0fxj8N6q2pBkJ3BPki8Dj9IEIzXz9XrfDyWZT3MEawi4/GAOUq+Ia3CNH0q+5xofWMcCP2hiiswG1lfVg0m24Gd1UuUxYkmSJEmSJE0/j9pJkiRJkiSpLww8SZIkSZIkqS8MPEmSJEmSJKkvDDxJkiRJkiSpLww8SZIkSZIkqS8MPEmSpIGR5N+d9PuSPJnk+DFlLkkykuT0Tt5jSRZM9xj6KcndSX6X5Opxnl3czmlHkp1JVrb565Isfxl9LUhy0XSMW5IkHVoMPEmSpIGT5D3A14Fzq+qpcYo8DXz+lR3VxJLMnmS51wFnVtXpVfXVMc/eC1wFLKuqU4C3AM8d4NAWAFMKPCWZdYB9SpKkAWDgSZIkDZQki4HbgfOq6o89im0ATknypnHqd3dNLU+yrk2vS3Jbkk1J/pTknCR3Jnl8tEyn3s1JtiX5WZL5bd6JSR5MsjXJL5Kc1Gl3bZJNwJox7RyR5NtJhpM8muRd7aONwGuTDLXz7boOWFlVzwBU1QtVdfs489yd5Jg2/bYkP2/T57TtDrV9zgNWA4vbvKuTzEpyU5It7a6ry9q6S9rfz3pgOMncJD9Msr3dgfXhHu9DkiQNqEn9V02SJGmGmAPcDyypqif2U24EuBFYBXxiCu2/Gng38H7gAeAs4JPAliSLqmoImAtsq6rPJfkicD3waeBbwOVV9WSSdwC3tm0BLASWVtWLY/q7EqCqTmsDVRuTLGz731BVi8YZ46nA1inMaayVwJVVtTnJUcALwLU0wazzAZJ8Cniuqs5IMgfYnGRjW//twKlVtSvJBcAzVXVeW+9VBzAuSZI0A7njSZIkDZL/Ar8CLp1E2fXAO5OcMIX2H6iqAoaBZ6tquKpGgB00x9GgCWp9v03fBZzdBnDOBO5LMgR8Eziu0+594wSdAM4GvgvQBtL+TBOk6qfNwNoknwWOrqp945RZBlzczuUR4DXAG9tnv6mqXW16GFiaZE2SxVV1oEf+JEnSDGPgSZIkDZIR4EPAGUlW7a9gG1C5Gbhm7KNO+ogxz/Z2+tnbyR+h907yovnM9Y+qWtT5OblTZk+Puuk1/v3YAbx1EuX28dJnwf/Ps6pW0+ziOhL49eiRwHHG9ZnOXE6oqtEdT3s6bf2hHcswcEO7A0ySJB1CDDxJkqSBUlXPA+cDK5JMtPNpHbAUmN/JezbJyUkOAz74MoZwGDD6zXEXAb+sqn8Cu5JcCJDGmyfR1sPAirbOQuB44PcT1LkBuLG9gJwkc9rdS2Pt5qUA1QWjmUlObHdyrQF+C5wE/AuY16n7E+CKJIePji3J3LEdJHk98HxV3QV8heaic0mSdAjxjidJkjRwqurvSc4FHk7yt6q6v0e5/yT5GnBLJ/tamsvH/wI8Bhw1xe730FxcvpXm2+RGL9ReAdyW5AvA4cA9wPYJ2roV+EaSYZodSpdU1d6k90aoqvpRkmOBn6YpWMCd4xT9EnBHuzPskU7+Ve0l5i8CO4Ef0+zo2pdkO02w7haao4Xb2j7+CnxgnD5OA25KMkJzDPKKCeYrSZIGTJprCiRJkiRJkqTp5VE7SZIkSZIk9YWBJ0mSJEmSJPWFgSdJkiRJkiT1hYEnSZIkSZIk9YWBJ0mSJEmSJPWFgSdJkiRJkiT1hYEnSZIkSZIk9YWBJ0mSJEmSJPXF/wA8MCSei4+nBQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize = (20, 10))\n",
    "\n",
    "# creating the bar plot \n",
    "plt.bar(score['Cluster Number'],score['Inertia']) \n",
    "\n",
    "plt.xlabel(\"K Number of Clusters\") \n",
    "plt.ylabel(\"Inertia\") \n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Silhouette Coefficient\n",
    "-1 to 1 | Higher value is better"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize = (20, 10)) \n",
    "  \n",
    "# creating the bar plot \n",
    "plt.bar(score['Cluster Number'],score['Silhouette Coefficient']) \n",
    "  \n",
    "plt.xlabel(\"K Number of Clusters\") \n",
    "plt.ylabel(\"Silhouette Coefficient\") \n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Calinski-Harabasz Index"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Higher value is better"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize = (20, 10)) \n",
    "  \n",
    "# creating the bar plot \n",
    "plt.bar(score['Cluster Number'],score['Calinski-Harabasz Index']) \n",
    "  \n",
    "plt.xlabel(\"K Number of Clusters\") \n",
    "plt.ylabel(\"Calinski-Harabasz Index\") \n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Davies-Bouldin Index"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Lower value is better"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize = (20, 10)) \n",
    "  \n",
    "# creating the bar plot \n",
    "plt.bar(score['Cluster Number'],score['Davies-Bouldin']) \n",
    "  \n",
    "plt.xlabel(\"K Number of Clusters\") \n",
    "plt.ylabel(\"Davies-Bouldin  Index\") \n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Recommendation using K Means with k=42"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Wall time: 5.39 s\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "KMeans(algorithm='auto', copy_x=True, init='k-means++', max_iter=300,\n",
       "       n_clusters=42, n_init=10, n_jobs=None, precompute_distances='auto',\n",
       "       random_state=None, tol=0.0001, verbose=0)"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%%time\n",
    "kmeans = KMeans(n_clusters = 42, init='k-means++')\n",
    "kmeans.fit(anime)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Inertia: 30588.067305598615\n",
      "Silhouette Coefficent: 0.14555913750633626\n",
      "Calinski Harabasz Index: 358.8252411316751\n",
      "Davies Bouldin Index: 2.2403835266604575\n"
     ]
    }
   ],
   "source": [
    "print('Inertia:', kmeans.inertia_ )\n",
    "print('Silhouette Coefficent:',metrics.silhouette_score(anime,kmeans.labels_, metric='euclidean'))\n",
    "print('Calinski Harabasz Index:',metrics.calinski_harabasz_score(anime, kmeans.labels_))\n",
    "print('Davies Bouldin Index:',davies_bouldin_score(anime, kmeans.labels_))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [],
   "source": [
    "df1 = df\n",
    "df1['label'] = kmeans.labels_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_recommendation_k(name):\n",
    "    rec_id=df1.index[df1['name']==name].tolist()[0]\n",
    "    rec_label = df1.iloc[rec_id]['label']\n",
    "    df2=df1[df1['label'] == rec_label].head(10)\n",
    "    print(\"Anime Recommendation for:\", name, '/',df.iloc[rec_id]['title_english'])\n",
    "    display(df2[['name','title_english','type','rating','genre']])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Anime Recommendation for: Kimi no Na wa. / Your Name.\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>name</th>\n",
       "      <th>title_english</th>\n",
       "      <th>type</th>\n",
       "      <th>rating</th>\n",
       "      <th>genre</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>254</td>\n",
       "      <td>Ace wo Nerae! (1979)</td>\n",
       "      <td>Aim for the Ace! (1979)</td>\n",
       "      <td>Movie</td>\n",
       "      <td>6.859385</td>\n",
       "      <td>Sports, Drama, Romance, School, Shoujo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>343</td>\n",
       "      <td>Blood: The Last Vampire</td>\n",
       "      <td>Blood: The Last Vampire</td>\n",
       "      <td>Movie</td>\n",
       "      <td>6.969985</td>\n",
       "      <td>Horror, Supernatural, Vampire</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>375</td>\n",
       "      <td>Shoujo Kakumei Utena: Adolescence Mokushiroku</td>\n",
       "      <td>Revolutionary Girl Utena: The Adolescence of U...</td>\n",
       "      <td>Movie</td>\n",
       "      <td>7.599917</td>\n",
       "      <td>Dementia, Drama, Fantasy, Romance, Shoujo, Sho...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>408</td>\n",
       "      <td>Hotori: Tada Saiwai wo Koinegau</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Movie</td>\n",
       "      <td>7.049512</td>\n",
       "      <td>Drama, Sci-Fi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>444</td>\n",
       "      <td>Marmalade Boy Movie</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Movie</td>\n",
       "      <td>7.469631</td>\n",
       "      <td>Romance, Shoujo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>615</td>\n",
       "      <td>Candy Candy (Movie)</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Movie</td>\n",
       "      <td>6.759582</td>\n",
       "      <td>Slice of Life, Drama, Shoujo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>932</td>\n",
       "      <td>Kidou Keisatsu Patlabor the Movie</td>\n",
       "      <td>Patlabor: The Movie</td>\n",
       "      <td>Movie</td>\n",
       "      <td>7.559880</td>\n",
       "      <td>Military, Police, Drama, Mecha</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1025</td>\n",
       "      <td>Nitaboh</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Movie</td>\n",
       "      <td>7.469480</td>\n",
       "      <td>Historical, Music</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1028</td>\n",
       "      <td>Ihatov Gensou: Kenji no Haru</td>\n",
       "      <td>Spring and Chaos</td>\n",
       "      <td>Movie</td>\n",
       "      <td>7.319390</td>\n",
       "      <td>Drama, Historical, Slice of Life</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1277</td>\n",
       "      <td>Aru Machi Kado no Monogatari</td>\n",
       "      <td>Tales of the Street Corner</td>\n",
       "      <td>Movie</td>\n",
       "      <td>6.679634</td>\n",
       "      <td>Drama, Music, Romance</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                               name  \\\n",
       "254                            Ace wo Nerae! (1979)   \n",
       "343                         Blood: The Last Vampire   \n",
       "375   Shoujo Kakumei Utena: Adolescence Mokushiroku   \n",
       "408                 Hotori: Tada Saiwai wo Koinegau   \n",
       "444                             Marmalade Boy Movie   \n",
       "615                             Candy Candy (Movie)   \n",
       "932               Kidou Keisatsu Patlabor the Movie   \n",
       "1025                                        Nitaboh   \n",
       "1028                   Ihatov Gensou: Kenji no Haru   \n",
       "1277                   Aru Machi Kado no Monogatari   \n",
       "\n",
       "                                          title_english   type    rating  \\\n",
       "254                             Aim for the Ace! (1979)  Movie  6.859385   \n",
       "343                             Blood: The Last Vampire  Movie  6.969985   \n",
       "375   Revolutionary Girl Utena: The Adolescence of U...  Movie  7.599917   \n",
       "408                                                 NaN  Movie  7.049512   \n",
       "444                                                 NaN  Movie  7.469631   \n",
       "615                                                 NaN  Movie  6.759582   \n",
       "932                                 Patlabor: The Movie  Movie  7.559880   \n",
       "1025                                                NaN  Movie  7.469480   \n",
       "1028                                   Spring and Chaos  Movie  7.319390   \n",
       "1277                         Tales of the Street Corner  Movie  6.679634   \n",
       "\n",
       "                                                  genre  \n",
       "254              Sports, Drama, Romance, School, Shoujo  \n",
       "343                       Horror, Supernatural, Vampire  \n",
       "375   Dementia, Drama, Fantasy, Romance, Shoujo, Sho...  \n",
       "408                                       Drama, Sci-Fi  \n",
       "444                                     Romance, Shoujo  \n",
       "615                        Slice of Life, Drama, Shoujo  \n",
       "932                      Military, Police, Drama, Mecha  \n",
       "1025                                  Historical, Music  \n",
       "1028                   Drama, Historical, Slice of Life  \n",
       "1277                              Drama, Music, Romance  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "get_recommendation_k('Kimi no Na wa.')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}