main.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from bayes import BayesianFilter\n",
    "bf = BayesianFilter()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# テキストを学習\n",
    "bf.fit(\"Adam Smith\", \"外国人\")\n",
    "bf.fit(\"Napoleon Bonaparte\", \"外国人\")\n",
    "bf.fit(\"Adolf Hitler\", \"外国人\")\n",
    "bf.fit(\"Gabriel Lippmann\", \"外国人\")\n",
    "bf.fit(\"トーマス　ベイズ\", \"カタカナ外国人\")\n",
    "bf.fit(\"カール　ハイド\",\"カタカナ外国人\")\n",
    "bf.fit(\"マーク　ザッカーバーグ\", \"カタカナ外国人\")\n",
    "bf.fit(\"リー　クワンユー\", \"カタカナ外国人\")\n",
    "bf.fit(\"湯川　秀樹\", \"日本人\")\n",
    "bf.fit(\"朝永　振一郎\",\"日本人\")\n",
    "bf.fit(\"小林　誠\",\"日本人\")\n",
    "bf.fit(\"益川　敏英\",\"日本人\")\n",
    "bf.fit(\"毛　沢東\", \"漢字外国人\")\n",
    "bf.fit(\"習　近平\", \"漢字外国人\")\n",
    "bf.fit(\"金　日成\", \"漢字外国人\")\n",
    "bf.fit(\"江　沢民\", \"漢字外国人\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "カタカナ外国人 [('外国人', -12.936737166250065), ('カタカナ外国人', -11.327299253815966), ('日本人', -11.390459481409463), ('漢字外国人', -11.327299253815966)]\n"
     ]
    }
   ],
   "source": [
    "# 予測\n",
    "pre, scorelist = bf.predict(\"近衛　文麿\")\n",
    "print(pre, scorelist)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{' ',\n",
       " 'Adam',\n",
       " 'Adolf',\n",
       " 'Bonaparte',\n",
       " 'Gabriel',\n",
       " 'Hitler',\n",
       " 'Lippmann',\n",
       " 'Napoleon',\n",
       " 'Smith',\n",
       " '\\u3000',\n",
       " 'カール',\n",
       " 'クワンユー',\n",
       " 'ザッカーバーグ',\n",
       " 'トーマス',\n",
       " 'ハイド',\n",
       " 'ベイズ',\n",
       " 'マーク',\n",
       " 'リー',\n",
       " '小林',\n",
       " '川',\n",
       " '振一郎',\n",
       " '敏英',\n",
       " '日成',\n",
       " '朝永',\n",
       " '毛',\n",
       " '江',\n",
       " '沢東',\n",
       " '沢民',\n",
       " '湯川',\n",
       " '益',\n",
       " '秀樹',\n",
       " '習',\n",
       " '誠',\n",
       " '近平',\n",
       " '金'}"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 出現した単語を全て記録したセットを取得\n",
    "wos = bf.words\n",
    "wosd = dict.fromkeys(wos, 0)\n",
    "wos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'カタカナ外国人': {'\\u3000': 4,\n",
       "  'カール': 1,\n",
       "  'クワンユー': 1,\n",
       "  'ザッカーバーグ': 1,\n",
       "  'トーマス': 1,\n",
       "  'ハイド': 1,\n",
       "  'ベイズ': 1,\n",
       "  'マーク': 1,\n",
       "  'リー': 1},\n",
       " '外国人': {' ': 4,\n",
       "  'Adam': 1,\n",
       "  'Adolf': 1,\n",
       "  'Bonaparte': 1,\n",
       "  'Gabriel': 1,\n",
       "  'Hitler': 1,\n",
       "  'Lippmann': 1,\n",
       "  'Napoleon': 1,\n",
       "  'Smith': 1},\n",
       " '日本人': {'\\u3000': 4,\n",
       "  '小林': 1,\n",
       "  '川': 1,\n",
       "  '振一郎': 1,\n",
       "  '敏英': 1,\n",
       "  '朝永': 1,\n",
       "  '湯川': 1,\n",
       "  '益': 1,\n",
       "  '秀樹': 1,\n",
       "  '誠': 1},\n",
       " '漢字外国人': {'\\u3000': 4,\n",
       "  '日成': 1,\n",
       "  '毛': 1,\n",
       "  '江': 1,\n",
       "  '沢東': 1,\n",
       "  '沢民': 1,\n",
       "  '習': 1,\n",
       "  '近平': 1,\n",
       "  '金': 1}}"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# カテゴリごとの単語出現回数を記録した辞書を取得\n",
    "wod = bf.word_dict\n",
    "wod"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'カタカナ外国人': 4, '外国人': 4, '日本人': 4, '漢字外国人': 4}"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# カテゴリの出現回数を記録した辞書を取得\n",
    "cad = bf.category_dict\n",
    "cad"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import json"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# convert dict object to json file\n",
    "with open('wosd.json', 'w') as fp:\n",
    "    json.dump(wosd, fp)\n",
    "\n",
    "with open('wod.json', 'w') as fp:\n",
    "    json.dump(wod, fp)\n",
    "    \n",
    "with open('cad.json', 'w') as fp:\n",
    "    json.dump(cad, fp)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# create dict object from json file\n",
    "wosd = json.loads(open('wosd.json').read())\n",
    "wos = set(wosd.keys())\n",
    "\n",
    "wod = json.loads(open('wod.json').read())\n",
    "cad = json.loads(open('cad.json').read())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# 一度初期化し、保存した過去の学習情報を引き継ぐ\n",
    "from bayes import BayesianFilter\n",
    "bf = BayesianFilter()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "set()"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bf.words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{}"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bf.word_dict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{}"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bf.category_dict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "bf.words = wos\n",
    "bf.word_dict = wod\n",
    "bf.category_dict = cad"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('外国人',\n",
       " [('外国人', -11.327299253815966),\n",
       "  ('カタカナ外国人', -12.936737166250065),\n",
       "  ('日本人', -12.999897393843563),\n",
       "  ('漢字外国人', -12.936737166250065)])"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bf.predict(\"Tom Hanks\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('外国人',\n",
       " [('外国人', -17.41815654480198),\n",
       "  ('カタカナ外国人', -20.637032369670184),\n",
       "  ('日本人', -20.742299415659346),\n",
       "  ('漢字外国人', -20.637032369670184)])"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bf.predict(\"Robert De Niro\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('外国人',\n",
       " [('外国人', -11.327299253815966),\n",
       "  ('カタカナ外国人', -12.936737166250065),\n",
       "  ('日本人', -12.999897393843563),\n",
       "  ('漢字外国人', -12.936737166250065)])"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bf.predict(\"Gen Hoshino\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('漢字外国人',\n",
       " [('外国人', -12.936737166250065),\n",
       "  ('カタカナ外国人', -11.327299253815966),\n",
       "  ('日本人', -11.390459481409463),\n",
       "  ('漢字外国人', -10.634152073256022)])"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bf.predict(\"金　正男\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('カタカナ外国人',\n",
       " [('外国人', -12.936737166250065),\n",
       "  ('カタカナ外国人', -11.327299253815966),\n",
       "  ('日本人', -11.390459481409463),\n",
       "  ('漢字外国人', -11.327299253815966)])"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bf.predict(\"朴　璐美\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('カタカナ外国人',\n",
       " [('外国人', -16.786884767960125),\n",
       "  ('カタカナ外国人', -15.177446855526025),\n",
       "  ('日本人', -15.261660492317354),\n",
       "  ('漢字外国人', -15.177446855526025)])"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bf.predict(\"李　小龍\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('カタカナ外国人',\n",
       " [('外国人', -20.637032369670184),\n",
       "  ('カタカナ外国人', -19.027594457236084),\n",
       "  ('日本人', -19.132861503225246),\n",
       "  ('漢字外国人', -19.027594457236084)])"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bf.predict(\"古歩道　ベンジャミン\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('カタカナ外国人',\n",
       " [('外国人', -12.936737166250065),\n",
       "  ('カタカナ外国人', -11.327299253815966),\n",
       "  ('日本人', -11.390459481409463),\n",
       "  ('漢字外国人', -11.327299253815966)])"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bf.predict(\"キム　イルソン\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('外国人',\n",
       " [('外国人', -11.327299253815966),\n",
       "  ('カタカナ外国人', -12.936737166250065),\n",
       "  ('日本人', -12.999897393843563),\n",
       "  ('漢字外国人', -12.936737166250065)])"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bf.predict(\"山下 奉文\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('カタカナ外国人',\n",
       " [('外国人', -12.936737166250065),\n",
       "  ('カタカナ外国人', -11.327299253815966),\n",
       "  ('日本人', -11.390459481409463),\n",
       "  ('漢字外国人', -11.327299253815966)])"
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bf.predict(\"宮沢　賢治\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('カタカナ外国人',\n",
       " [('外国人', -12.936737166250065),\n",
       "  ('カタカナ外国人', -11.327299253815966),\n",
       "  ('日本人', -11.390459481409463),\n",
       "  ('漢字外国人', -11.327299253815966)])"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bf.predict(\"徳川　家康\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('カタカナ外国人',\n",
       " [('外国人', -12.936737166250065),\n",
       "  ('カタカナ外国人', -11.327299253815966),\n",
       "  ('日本人', -11.390459481409463),\n",
       "  ('漢字外国人', -11.327299253815966)])"
      ]
     },
     "execution_count": 93,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bf.predict(\"井浦　新\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('カタカナ外国人',\n",
       " [('外国人', -16.786884767960125),\n",
       "  ('カタカナ外国人', -15.177446855526025),\n",
       "  ('日本人', -15.261660492317354),\n",
       "  ('漢字外国人', -15.177446855526025)])"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bf.predict(\"窪塚　洋介\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('カタカナ外国人',\n",
       " [('外国人', -12.936737166250065),\n",
       "  ('カタカナ外国人', -11.327299253815966),\n",
       "  ('日本人', -11.390459481409463),\n",
       "  ('漢字外国人', -11.327299253815966)])"
      ]
     },
     "execution_count": 95,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bf.predict(\"伊藤　博文\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "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
}