{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "import numpy as np\n",
    "np.set_printoptions(precision=2)\n",
    "tfidf = TfidfVectorizer(token_pattern=u'(?u)\\\\b\\\\w+\\\\b')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "result = tfidf.fit_transform([\n",
    "    '亀 亀 世界',\n",
    "    'うさぎ 小山 亀',\n",
    "    '桃太郎 桃太郎 腰 きびだんご']) # うさぎとかめの歌と桃太郎の歌（名刺のみ）を学習"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'うさぎ': 0, 'きびだんご': 1, '世界': 2, '亀': 3, '小山': 4, '桃太郎': 5, '腰': 6}"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tfidf.vocabulary_  # この学習によって取り込んだ語彙"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.  , 0.  , 0.55, 0.84, 0.  , 0.  , 0.  ],\n",
       "       [0.62, 0.  , 0.  , 0.47, 0.62, 0.  , 0.  ],\n",
       "       [0.  , 0.41, 0.  , 0.  , 0.  , 0.82, 0.41]])"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result.toarray()  # TFIDF値"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.  , 0.  , 0.55, 0.84, 0.  , 0.  , 0.  ],\n",
       "       [0.62, 0.  , 0.  , 0.47, 0.62, 0.  , 0.  ]])"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "usagi_arrays = result.toarray()[[True, True, False]]  # うさぎとかめの歌だけ抽出\n",
    "usagi_arrays"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.31, 0.  , 0.27, 0.65, 0.31, 0.  , 0.  ])"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "usagi_weight = np.mean(usagi_arrays, axis=0)  # 重みを計算\n",
    "usagi_weight"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.  , 0.41, 0.  , 0.  , 0.  , 0.82, 0.41])"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "momotaro_arrays = result.toarray()[[False, False, True]]  # ももたろうの歌だけ抽出\n",
    "momotaro_arrays\n",
    "momotaro_weight = np.mean(momotaro_arrays, axis=0)  # 重みを計算\n",
    "momotaro_weight"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0., 0., 0., 1., 0., 0., 0.]])"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_song = tfidf.transform(['浦島 亀 竜宮城 絵']).toarray() # 新たな歌が与えられた\n",
    "new_song"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.65])"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "score = new_song.dot(usagi_weight)\n",
    "score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "score = new_song.dot(momotaro_weight)\n",
    "score"
   ]
  },
  {
   "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": "3.6.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}