0502_LSA_LDA.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 0. 데이터로딩"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dict_keys(['cv', 'tfidf', 'tdm_cv', 'tdm_tfidf'])\n"
     ]
    }
   ],
   "source": [
    "import joblib\n",
    "import pandas as pd\n",
    "def get_nouns(text):\n",
    "    return 1\n",
    "with open('tdm.pkl', 'rb') as f:\n",
    "    data = joblib.load(f)\n",
    "    print(data.keys())\n",
    "    locals().update(data)\n",
    "    del data\n",
    "df = pd.read_csv('review_full.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. LSA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(36477, 1000)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tdm_tfidf.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.decomposition import TruncatedSVD\n",
    "svd = TruncatedSVD(n_components=30)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(36477, 30)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pos = svd.fit_transform(tdm_tfidf)\n",
    "pos.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "5.0    16335\n",
       "4.0    13131\n",
       "3.0     4814\n",
       "2.0     1166\n",
       "1.0      864\n",
       "0.0      167\n",
       "Name: 평점, dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.평점.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1d1abc65a30>]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "dim0 = 0\n",
    "dim1 = 1\n",
    "plt.plot(pos[df.평점==1, dim0], pos[df.평점==1, dim1], 'bo') # 1점은 파란색\n",
    "plt.plot(pos[df.평점==5, dim0], pos[df.평점==5, dim1], 'ro') # 5점은 빨간색"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import Normalizer\n",
    "norm = Normalizer(copy=False)\n",
    "pos2 = norm.fit_transform(pos)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1d1abd737c0>]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "dim0 = 0\n",
    "dim1 = 1\n",
    "plt.plot(pos2[df.평점==1, dim0], pos2[df.평점==1, dim1], 'bo') # 1점은 파란색\n",
    "plt.plot(pos2[df.평점==5, dim0], pos2[df.평점==5, dim1], 'ro') # 5점은 빨간색"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.pipeline import make_pipeline\n",
    "svd = TruncatedSVD(n_components=30)\n",
    "norm = Normalizer(copy=False)\n",
    "lsa = make_pipeline(svd, norm)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "pos3 = lsa.fit_transform(tdm_cv)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1d1abe23850>]"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "dim0 = 0\n",
    "dim1 = 1\n",
    "plt.plot(pos3[df.평점==1, dim0], pos3[df.평점==1, dim1], 'bo') # 1점은 파란색\n",
    "plt.plot(pos3[df.평점==5, dim0], pos3[df.평점==5, dim1], 'ro') # 5점은 빨간색"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2. LDA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "words = cv.get_feature_names()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{0: '!!',\n",
       " 1: '!!!',\n",
       " 2: '가게',\n",
       " 3: '가격',\n",
       " 4: '가도',\n",
       " 5: '가루',\n",
       " 6: '가리비',\n",
       " 7: '가면',\n",
       " 8: '가보',\n",
       " 9: '가시',\n",
       " 10: '가야',\n",
       " 11: '가이드',\n",
       " 12: '가정',\n",
       " 13: '가족',\n",
       " 14: '가지',\n",
       " 15: '가짓수',\n",
       " 16: '가치',\n",
       " 17: '각종',\n",
       " 18: '간격',\n",
       " 19: '간식',\n",
       " 20: '간장',\n",
       " 21: '간판',\n",
       " 22: '갈비',\n",
       " 23: '갈비탕',\n",
       " 24: '감동',\n",
       " 25: '감사',\n",
       " 26: '감안',\n",
       " 27: '감자',\n",
       " 28: '감자 튀김',\n",
       " 29: '감자탕',\n",
       " 30: '강남',\n",
       " 31: '개인',\n",
       " 32: '객실',\n",
       " 33: '거리',\n",
       " 34: '걱정',\n",
       " 35: '건강',\n",
       " 36: '건너편',\n",
       " 37: '건담',\n",
       " 38: '건물',\n",
       " 39: '건지',\n",
       " 40: '검색',\n",
       " 41: '게장',\n",
       " 42: '겨울',\n",
       " 43: '결정',\n",
       " 44: '경우',\n",
       " 45: '경주',\n",
       " 46: '경치',\n",
       " 47: '경험',\n",
       " 48: '계단',\n",
       " 49: '계란',\n",
       " 50: '계산',\n",
       " 51: '계속',\n",
       " 52: '계절',\n",
       " 53: '계획',\n",
       " 54: '고객',\n",
       " 55: '고급',\n",
       " 56: '고기',\n",
       " 57: '고기국수',\n",
       " 58: '고려',\n",
       " 59: '고민',\n",
       " 60: '고베',\n",
       " 61: '고추',\n",
       " 62: '고추장',\n",
       " 63: '골목',\n",
       " 64: '곰탕',\n",
       " 65: '곱창',\n",
       " 66: '곳도',\n",
       " 67: '공간',\n",
       " 68: '공기',\n",
       " 69: '공연',\n",
       " 70: '공유',\n",
       " 71: '공항',\n",
       " 72: '과일',\n",
       " 73: '과자',\n",
       " 74: '과정',\n",
       " 75: '관광',\n",
       " 76: '관광객',\n",
       " 77: '관광지',\n",
       " 78: '관리',\n",
       " 79: '관심',\n",
       " 80: '광고',\n",
       " 81: '광안리',\n",
       " 82: '광화문',\n",
       " 83: '구경',\n",
       " 84: '구매',\n",
       " 85: '구사',\n",
       " 86: '구성',\n",
       " 87: '구이',\n",
       " 88: '구입',\n",
       " 89: '국물',\n",
       " 90: '국밥',\n",
       " 91: '국수',\n",
       " 92: '국제',\n",
       " 93: '군데',\n",
       " 94: '군만두',\n",
       " 95: '권장',\n",
       " 96: '규모',\n",
       " 97: '그날',\n",
       " 98: '그니',\n",
       " 99: '그때',\n",
       " 100: '그런지',\n",
       " 101: '그룹',\n",
       " 102: '그릇',\n",
       " 103: '그릴',\n",
       " 104: '그중',\n",
       " 105: '근처',\n",
       " 106: '금액',\n",
       " 107: '금요일',\n",
       " 108: '기간',\n",
       " 109: '기념일',\n",
       " 110: '기대',\n",
       " 111: '기름',\n",
       " 112: '기본',\n",
       " 113: '기분',\n",
       " 114: '기억',\n",
       " 115: '기준',\n",
       " 116: '기차',\n",
       " 117: '기타',\n",
       " 118: '기회',\n",
       " 119: '긴자',\n",
       " 120: '김밥',\n",
       " 121: '김치',\n",
       " 122: '김치찌개',\n",
       " 123: '깍두기',\n",
       " 124: '껍질',\n",
       " 125: '꼬치',\n",
       " 126: '나라',\n",
       " 127: '나름',\n",
       " 128: '나머지',\n",
       " 129: '나무',\n",
       " 130: '나중',\n",
       " 131: '낙지',\n",
       " 132: '날씨',\n",
       " 133: '남자',\n",
       " 134: '남자친구',\n",
       " 135: '남편',\n",
       " 136: '남포동',\n",
       " 137: '내부',\n",
       " 138: '내장',\n",
       " 139: '냄비',\n",
       " 140: '냄새',\n",
       " 141: '냉면',\n",
       " 142: '노래',\n",
       " 143: '노력',\n",
       " 144: '녹차',\n",
       " 145: '뉴욕',\n",
       " 146: '느낌',\n",
       " 147: '니다',\n",
       " 148: '닌자',\n",
       " 149: '다리',\n",
       " 150: '다음',\n",
       " 151: '다음날',\n",
       " 152: '다음번',\n",
       " 153: '다행',\n",
       " 154: '단골',\n",
       " 155: '단맛',\n",
       " 156: '단위',\n",
       " 157: '단점',\n",
       " 158: '단체',\n",
       " 159: '단팥빵',\n",
       " 160: '달걀',\n",
       " 161: '달러',\n",
       " 162: '닭갈비',\n",
       " 163: '닭강정',\n",
       " 164: '닭고기',\n",
       " 165: '대가',\n",
       " 166: '대게',\n",
       " 167: '대기',\n",
       " 168: '대도',\n",
       " 169: '대로',\n",
       " 170: '대부분',\n",
       " 171: '대비',\n",
       " 172: '대신',\n",
       " 173: '대전',\n",
       " 174: '대전역',\n",
       " 175: '대접',\n",
       " 176: '대중',\n",
       " 177: '대체',\n",
       " 178: '대표',\n",
       " 179: '대형',\n",
       " 180: '대화',\n",
       " 181: '덕분',\n",
       " 182: '데이트',\n",
       " 183: '도넛',\n",
       " 184: '도로',\n",
       " 185: '도보',\n",
       " 186: '도시',\n",
       " 187: '도우',\n",
       " 188: '도움',\n",
       " 189: '도전',\n",
       " 190: '도착',\n",
       " 191: '도쿄',\n",
       " 192: '도쿄 역',\n",
       " 193: '동네',\n",
       " 194: '동대문',\n",
       " 195: '동료',\n",
       " 196: '동시',\n",
       " 197: '동안',\n",
       " 198: '돼지',\n",
       " 199: '돼지고기',\n",
       " 200: '돼지국밥',\n",
       " 201: '된장',\n",
       " 202: '된장국',\n",
       " 203: '된장찌개',\n",
       " 204: '두부',\n",
       " 205: '둔산동',\n",
       " 206: '등급',\n",
       " 207: '등등',\n",
       " 208: '등심',\n",
       " 209: '디너',\n",
       " 210: '디자인',\n",
       " 211: '디저트',\n",
       " 212: '딤섬',\n",
       " 213: '딸기',\n",
       " 214: '때문',\n",
       " 215: '떡볶이',\n",
       " 216: '라고',\n",
       " 217: '라멘',\n",
       " 218: '라면',\n",
       " 219: '라운지',\n",
       " 220: '라이브',\n",
       " 221: '라인',\n",
       " 222: '라자냐',\n",
       " 223: '랜드',\n",
       " 224: '레드',\n",
       " 225: '레스토랑',\n",
       " 226: '로비',\n",
       " 227: '로컬',\n",
       " 228: '롯데',\n",
       " 229: '롯데리아',\n",
       " 230: '롯데호텔',\n",
       " 231: '리뷰',\n",
       " 232: '리스트',\n",
       " 233: '리필',\n",
       " 234: '마늘',\n",
       " 235: '마다',\n",
       " 236: '마리',\n",
       " 237: '마무리',\n",
       " 238: '마시기',\n",
       " 239: '마을',\n",
       " 240: '마음',\n",
       " 241: '마지막',\n",
       " 242: '마찬가지',\n",
       " 243: '마카롱',\n",
       " 244: '막걸리',\n",
       " 245: '막국수',\n",
       " 246: '만두',\n",
       " 247: '만석',\n",
       " 248: '만원',\n",
       " 249: '만족',\n",
       " 250: '만족도',\n",
       " 251: '만큼',\n",
       " 252: '말씀',\n",
       " 253: '맛나',\n",
       " 254: '맛집',\n",
       " 255: '망고',\n",
       " 256: '맞은편',\n",
       " 257: '매니저',\n",
       " 258: '매력',\n",
       " 259: '매운맛',\n",
       " 260: '매장',\n",
       " 261: '매콤',\n",
       " 262: '맥도날드',\n",
       " 263: '맥주',\n",
       " 264: '메뉴',\n",
       " 265: '메밀',\n",
       " 266: '메이드',\n",
       " 267: '메이드 카페',\n",
       " 268: '메인',\n",
       " 269: '멕시코',\n",
       " 270: '명동',\n",
       " 271: '명성',\n",
       " 272: '모델링',\n",
       " 273: '모두',\n",
       " 274: '모듬',\n",
       " 275: '모모',\n",
       " 276: '모습',\n",
       " 277: '모양',\n",
       " 278: '모임',\n",
       " 279: '목록',\n",
       " 280: '무례',\n",
       " 281: '무료',\n",
       " 282: '무시',\n",
       " 283: '무침',\n",
       " 284: '무한',\n",
       " 285: '문제',\n",
       " 286: '문화',\n",
       " 287: '물건',\n",
       " 288: '물고기',\n",
       " 289: '미국',\n",
       " 290: '미소',\n",
       " 291: '미역국',\n",
       " 292: '밀면',\n",
       " 293: '밑반찬',\n",
       " 294: '바다',\n",
       " 295: '바닥',\n",
       " 296: '바닷가',\n",
       " 297: '바람',\n",
       " 298: '바비큐',\n",
       " 299: '바텐더',\n",
       " 300: '반찬',\n",
       " 301: '발견',\n",
       " 302: '발리',\n",
       " 303: '방문',\n",
       " 304: '방법',\n",
       " 305: '방송',\n",
       " 306: '방식',\n",
       " 307: '방콕',\n",
       " 308: '배달',\n",
       " 309: '배치',\n",
       " 310: '백화점',\n",
       " 311: '밴드',\n",
       " 312: '버거',\n",
       " 313: '버섯',\n",
       " 314: '버스',\n",
       " 315: '버터',\n",
       " 316: '번호',\n",
       " 317: '베네딕트',\n",
       " 318: '베이커리',\n",
       " 319: '베이컨',\n",
       " 320: '베트남',\n",
       " 321: '변경',\n",
       " 322: '변화',\n",
       " 323: '별도',\n",
       " 324: '별미',\n",
       " 325: '보니',\n",
       " 326: '보석',\n",
       " 327: '보쌈',\n",
       " 328: '보장',\n",
       " 329: '보통',\n",
       " 330: '볶음',\n",
       " 331: '볶음밥',\n",
       " 332: '본점',\n",
       " 333: '봉사',\n",
       " 334: '부담',\n",
       " 335: '부모',\n",
       " 336: '부분',\n",
       " 337: '부산',\n",
       " 338: '부산대',\n",
       " 339: '부산역',\n",
       " 340: '부엌',\n",
       " 341: '부위',\n",
       " 342: '부족',\n",
       " 343: '부추',\n",
       " 344: '부탁',\n",
       " 345: '부페',\n",
       " 346: '북적',\n",
       " 347: '분도',\n",
       " 348: '분위기',\n",
       " 349: '분이',\n",
       " 350: '분점',\n",
       " 351: '불고기',\n",
       " 352: '불만',\n",
       " 353: '불편',\n",
       " 354: '뷔페',\n",
       " 355: '브랜드',\n",
       " 356: '브런치',\n",
       " 357: '블로그',\n",
       " 358: '비교',\n",
       " 359: '비빔밥',\n",
       " 360: '비용',\n",
       " 361: '비즈니스',\n",
       " 362: '빈대떡',\n",
       " 363: '빙수',\n",
       " 364: '빵집',\n",
       " 365: '사과',\n",
       " 366: '사람',\n",
       " 367: '사랑',\n",
       " 368: '사리',\n",
       " 369: '사막',\n",
       " 370: '사서',\n",
       " 371: '사시미',\n",
       " 372: '사실',\n",
       " 373: '사용',\n",
       " 374: '사용자',\n",
       " 375: '사이',\n",
       " 376: '사이드',\n",
       " 377: '사이즈',\n",
       " 378: '사이트',\n",
       " 379: '사장',\n",
       " 380: '사전',\n",
       " 381: '사진',\n",
       " 382: '사항',\n",
       " 383: '산책',\n",
       " 384: '삼겹살',\n",
       " 385: '삼계탕',\n",
       " 386: '상대',\n",
       " 387: '상점',\n",
       " 388: '상태',\n",
       " 389: '새우',\n",
       " 390: '샌드위치',\n",
       " 391: '샐러드',\n",
       " 392: '생각',\n",
       " 393: '생맥주',\n",
       " 394: '생선',\n",
       " 395: '생일',\n",
       " 396: '샤브샤브',\n",
       " 397: '샴페인',\n",
       " 398: '서면',\n",
       " 399: '서버',\n",
       " 400: '서비스',\n",
       " 401: '서빙',\n",
       " 402: '서서',\n",
       " 403: '서양',\n",
       " 404: '서양식',\n",
       " 405: '서울',\n",
       " 406: '서울역',\n",
       " 407: '선물',\n",
       " 408: '선택',\n",
       " 409: '선호',\n",
       " 410: '설렁탕',\n",
       " 411: '설명',\n",
       " 412: '설빙',\n",
       " 413: '설정',\n",
       " 414: '성급',\n",
       " 415: '성비',\n",
       " 416: '성심당',\n",
       " 417: '세계',\n",
       " 418: '세상',\n",
       " 419: '세트',\n",
       " 420: '섹션',\n",
       " 421: '센터',\n",
       " 422: '셀프',\n",
       " 423: '셰프',\n",
       " 424: '소개',\n",
       " 425: '소고기',\n",
       " 426: '소금',\n",
       " 427: '소리',\n",
       " 428: '소문',\n",
       " 429: '소바',\n",
       " 430: '소스',\n",
       " 431: '소시지',\n",
       " 432: '소주',\n",
       " 433: '속도',\n",
       " 434: '손님',\n",
       " 435: '쇠고기',\n",
       " 436: '쇼핑',\n",
       " 437: '수도',\n",
       " 438: '수요미식회',\n",
       " 439: '수용',\n",
       " 440: '수육',\n",
       " 441: '수제',\n",
       " 442: '수준',\n",
       " 443: '수프',\n",
       " 444: '수행',\n",
       " 445: '숙박',\n",
       " 446: '숙소',\n",
       " 447: '순간',\n",
       " 448: '순대',\n",
       " 449: '술집',\n",
       " 450: '숯불',\n",
       " 451: '쉐프',\n",
       " 452: '슈퍼',\n",
       " 453: '스타',\n",
       " 454: '스타벅스',\n",
       " 455: '스타일',\n",
       " 456: '스테이크',\n",
       " 457: '스튜',\n",
       " 458: '스파게티',\n",
       " 459: '스페셜',\n",
       " 460: '스페인',\n",
       " 461: '스프',\n",
       " 462: '시간',\n",
       " 463: '시간대',\n",
       " 464: '시금치',\n",
       " 465: '시내',\n",
       " 466: '시도',\n",
       " 467: '시설',\n",
       " 468: '시스템',\n",
       " 469: '시식',\n",
       " 470: '시작',\n",
       " 471: '시장',\n",
       " 472: '시청',\n",
       " 473: '시험',\n",
       " 474: '식당',\n",
       " 475: '식사',\n",
       " 476: '신경',\n",
       " 477: '신속',\n",
       " 478: '실내',\n",
       " 479: '실망',\n",
       " 480: '실수',\n",
       " 481: '실제',\n",
       " 482: '실패',\n",
       " 483: '쌀국수',\n",
       " 484: '아기',\n",
       " 485: '아내',\n",
       " 486: '아니',\n",
       " 487: '아들',\n",
       " 488: '아래',\n",
       " 489: '아메리카노',\n",
       " 490: '아무것',\n",
       " 491: '아보카도',\n",
       " 492: '아쉬움',\n",
       " 493: '아시아',\n",
       " 494: '아웃',\n",
       " 495: '아웃백',\n",
       " 496: '아이',\n",
       " 497: '아이들',\n",
       " 498: '아이스',\n",
       " 499: '아이스크림',\n",
       " 500: '아주머니',\n",
       " 501: '아침',\n",
       " 502: '아키하바라',\n",
       " 503: '안나',\n",
       " 504: '안내',\n",
       " 505: '안심',\n",
       " 506: '안주',\n",
       " 507: '안쪽',\n",
       " 508: '애프터',\n",
       " 509: '야경',\n",
       " 510: '야외',\n",
       " 511: '야채',\n",
       " 512: '약간',\n",
       " 513: '약속',\n",
       " 514: '양고기',\n",
       " 515: '양념',\n",
       " 516: '양도',\n",
       " 517: '양배추',\n",
       " 518: '양식',\n",
       " 519: '양은',\n",
       " 520: '양이',\n",
       " 521: '양파',\n",
       " 522: '얘기',\n",
       " 523: '어른',\n",
       " 524: '어머니',\n",
       " 525: '어묵',\n",
       " 526: '언급',\n",
       " 527: '언어',\n",
       " 528: '얼마',\n",
       " 529: '얼음',\n",
       " 530: '엄마',\n",
       " 531: '에그',\n",
       " 532: '에선',\n",
       " 533: '여름',\n",
       " 534: '여성',\n",
       " 535: '여유',\n",
       " 536: '여자',\n",
       " 537: '여자친구',\n",
       " 538: '여행',\n",
       " 539: '여행객',\n",
       " 540: '여행자',\n",
       " 541: '역사',\n",
       " 542: '연결',\n",
       " 543: '연기',\n",
       " 544: '연어',\n",
       " 545: '연인',\n",
       " 546: '연주',\n",
       " 547: '영어',\n",
       " 548: '영업',\n",
       " 549: '영화',\n",
       " 550: '예상',\n",
       " 551: '예술',\n",
       " 552: '예약',\n",
       " 553: '예전',\n",
       " 554: '예정',\n",
       " 555: '옛날',\n",
       " 556: '오늘',\n",
       " 557: '오뎅',\n",
       " 558: '오랜만',\n",
       " 559: '오랫동안',\n",
       " 560: '오른쪽',\n",
       " 561: '오리',\n",
       " 562: '오믈렛',\n",
       " 563: '오전',\n",
       " 564: '오징어',\n",
       " 565: '오픈',\n",
       " 566: '오후',\n",
       " 567: '온라인',\n",
       " 568: '옵션',\n",
       " 569: '와서',\n",
       " 570: '와이파이',\n",
       " 571: '와인',\n",
       " 572: '와플',\n",
       " 573: '완벽',\n",
       " 574: '완전',\n",
       " 575: '외관',\n",
       " 576: '외국',\n",
       " 577: '외국인',\n",
       " 578: '외부',\n",
       " 579: '외식',\n",
       " 580: '왼쪽',\n",
       " 581: '요구',\n",
       " 582: '요금',\n",
       " 583: '요리',\n",
       " 584: '요리사',\n",
       " 585: '요즘',\n",
       " 586: '요청',\n",
       " 587: '우동',\n",
       " 588: '우리나라',\n",
       " 589: '우수',\n",
       " 590: '우유',\n",
       " 591: '운영',\n",
       " 592: '원래',\n",
       " 593: '원조',\n",
       " 594: '웨이',\n",
       " 595: '웨이터',\n",
       " 596: '위생',\n",
       " 597: '위층',\n",
       " 598: '위치',\n",
       " 599: '유럽',\n",
       " 600: '유로',\n",
       " 601: '유리',\n",
       " 602: '유명',\n",
       " 603: '유명세',\n",
       " 604: '유명한',\n",
       " 605: '유일한',\n",
       " 606: '유지',\n",
       " 607: '유형',\n",
       " 608: '육개장',\n",
       " 609: '육수',\n",
       " 610: '육전',\n",
       " 611: '육회',\n",
       " 612: '음료',\n",
       " 613: '음식',\n",
       " 614: '음식점',\n",
       " 615: '음악',\n",
       " 616: '응대',\n",
       " 617: '의미',\n",
       " 618: '의사',\n",
       " 619: '의심',\n",
       " 620: '의자',\n",
       " 621: '이고',\n",
       " 622: '이나',\n",
       " 623: '이다',\n",
       " 624: '이동',\n",
       " 625: '이랑',\n",
       " 626: '이름',\n",
       " 627: '이번',\n",
       " 628: '이벤트',\n",
       " 629: '이상',\n",
       " 630: '이서',\n",
       " 631: '이스',\n",
       " 632: '이야기',\n",
       " 633: '이예',\n",
       " 634: '이외',\n",
       " 635: '이용',\n",
       " 636: '이유',\n",
       " 637: '이전',\n",
       " 638: '이제',\n",
       " 639: '이집',\n",
       " 640: '이탈리아',\n",
       " 641: '이탈리안',\n",
       " 642: '이태리',\n",
       " 643: '이태원',\n",
       " 644: '이하',\n",
       " 645: '이해',\n",
       " 646: '이후',\n",
       " 647: '인근',\n",
       " 648: '인기',\n",
       " 649: '인당',\n",
       " 650: '인도',\n",
       " 651: '인분',\n",
       " 652: '인사',\n",
       " 653: '인사동',\n",
       " 654: '인삼',\n",
       " 655: '인상',\n",
       " 656: '인생',\n",
       " 657: '인정',\n",
       " 658: '인지',\n",
       " 659: '인터넷',\n",
       " 660: '인테리어',\n",
       " 661: '일반',\n",
       " 662: '일본',\n",
       " 663: '일본어',\n",
       " 664: '일본인',\n",
       " 665: '일부',\n",
       " 666: '일식',\n",
       " 667: '일요일',\n",
       " 668: '일품',\n",
       " 669: '일행',\n",
       " 670: '임페리얼',\n",
       " 671: '입구',\n",
       " 672: '입맛',\n",
       " 673: '입안',\n",
       " 674: '입장',\n",
       " 675: '있다',\n",
       " 676: '자극',\n",
       " 677: '자랑',\n",
       " 678: '자리',\n",
       " 679: '자신',\n",
       " 680: '자주',\n",
       " 681: '자체',\n",
       " 682: '작년',\n",
       " 683: '작성',\n",
       " 684: '작업',\n",
       " 685: '잘 먹었습니다',\n",
       " 686: '잘못',\n",
       " 687: '장사',\n",
       " 688: '장소',\n",
       " 689: '장식',\n",
       " 690: '장어',\n",
       " 691: '장점',\n",
       " 692: '재료',\n",
       " 693: '재미',\n",
       " 694: '재즈',\n",
       " 695: '저녁',\n",
       " 696: '적극',\n",
       " 697: '전골',\n",
       " 698: '전국',\n",
       " 699: '전망',\n",
       " 700: '전문',\n",
       " 701: '전문점',\n",
       " 702: '전반',\n",
       " 703: '전복',\n",
       " 704: '전복죽',\n",
       " 705: '전용',\n",
       " 706: '전주',\n",
       " 707: '전채',\n",
       " 708: '전체',\n",
       " 709: '전통',\n",
       " 710: '전형',\n",
       " 711: '전화',\n",
       " 712: '절대',\n",
       " 713: '절반',\n",
       " 714: '점심',\n",
       " 715: '점심때',\n",
       " 716: '점심시간',\n",
       " 717: '접근성',\n",
       " 718: '접시',\n",
       " 719: '정도',\n",
       " 720: '정말',\n",
       " 721: '정보',\n",
       " 722: '정성',\n",
       " 723: '정식',\n",
       " 724: '정신',\n",
       " 725: '정원',\n",
       " 726: '정통',\n",
       " 727: '제가',\n",
       " 728: '제공',\n",
       " 729: '제안',\n",
       " 730: '제외',\n",
       " 731: '제일',\n",
       " 732: '제주',\n",
       " 733: '제주도',\n",
       " 734: '제품',\n",
       " 735: '제한',\n",
       " 736: '조각',\n",
       " 737: '조개',\n",
       " 738: '조금',\n",
       " 739: '조리',\n",
       " 740: '조명',\n",
       " 741: '조미료',\n",
       " 742: '조식',\n",
       " 743: '조심',\n",
       " 744: '조합',\n",
       " 745: '조화',\n",
       " 746: '족발',\n",
       " 747: '종류',\n",
       " 748: '종업원',\n",
       " 749: '종일',\n",
       " 750: '좋아서',\n",
       " 751: '좌석',\n",
       " 752: '주말',\n",
       " 753: '주문',\n",
       " 754: '주민',\n",
       " 755: '주방',\n",
       " 756: '주방장',\n",
       " 757: '주변',\n",
       " 758: '주스',\n",
       " 759: '주위',\n",
       " 760: '주의',\n",
       " 761: '주인',\n",
       " 762: '주차',\n",
       " 763: '주차장',\n",
       " 764: '준비',\n",
       " 765: '중간',\n",
       " 766: '중국',\n",
       " 767: '중독',\n",
       " 768: '중식',\n",
       " 769: '중심',\n",
       " 770: '쥬스',\n",
       " 771: '즉석',\n",
       " 772: '지금',\n",
       " 773: '지도',\n",
       " 774: '지만',\n",
       " 775: '지방',\n",
       " 776: '지불',\n",
       " 777: '지역',\n",
       " 778: '지원',\n",
       " 779: '지인',\n",
       " 780: '지점',\n",
       " 781: '지하',\n",
       " 782: '지하철',\n",
       " 783: '직원',\n",
       " 784: '직장인',\n",
       " 785: '진짜',\n",
       " 786: '진한',\n",
       " 787: '질문',\n",
       " 788: '집으로',\n",
       " 789: '집중',\n",
       " 790: '짜장면',\n",
       " 791: '짬뽕',\n",
       " 792: '차이',\n",
       " 793: '참고',\n",
       " 794: '참치',\n",
       " 795: '창가',\n",
       " 796: '창문',\n",
       " 797: '채소',\n",
       " 798: '채식',\n",
       " 799: '채식주의',\n",
       " 800: '채식주의자',\n",
       " 801: '처리',\n",
       " 802: '처음',\n",
       " 803: '천원',\n",
       " 804: '천장',\n",
       " 805: '철판',\n",
       " 806: '청결',\n",
       " 807: '청소',\n",
       " 808: '체인',\n",
       " 809: '체인점',\n",
       " 810: '체크',\n",
       " 811: '초밥',\n",
       " 812: '초콜릿',\n",
       " 813: '최고',\n",
       " 814: '최근',\n",
       " 815: '최대',\n",
       " 816: '최상',\n",
       " 817: '최악',\n",
       " 818: '추가',\n",
       " 819: '추어탕',\n",
       " 820: '추억',\n",
       " 821: '추천',\n",
       " 822: '축하',\n",
       " 823: '출구',\n",
       " 824: '출장',\n",
       " 825: '취향',\n",
       " 826: '치즈',\n",
       " 827: '치킨',\n",
       " 828: '친구',\n",
       " 829: '친절',\n",
       " 830: '칠리',\n",
       " 831: '카드',\n",
       " 832: '카레',\n",
       " 833: '카운터',\n",
       " 834: '카페',\n",
       " 835: '칵테일',\n",
       " 836: '칼국수',\n",
       " 837: '캐주얼',\n",
       " 838: '커리',\n",
       " 839: '커서',\n",
       " 840: '커플',\n",
       " 841: '커피',\n",
       " 842: '커피숍',\n",
       " 843: '케이크',\n",
       " 844: '케익',\n",
       " 845: '코너',\n",
       " 846: '코스',\n",
       " 847: '코트',\n",
       " 848: '콩국수',\n",
       " 849: '콩나물',\n",
       " 850: '쿠폰',\n",
       " 851: '크기',\n",
       " 852: '크림',\n",
       " 853: '크림치즈',\n",
       " 854: '클럽',\n",
       " 855: '키친',\n",
       " 856: '타워',\n",
       " 857: '타이',\n",
       " 858: '타임',\n",
       " 859: '타코',\n",
       " 860: '타파스',\n",
       " 861: '탕수육',\n",
       " 862: '태국',\n",
       " 863: '태도',\n",
       " 864: '택시',\n",
       " 865: '테라스',\n",
       " 866: '테마',\n",
       " 867: '테이블',\n",
       " 868: '테이크',\n",
       " 869: '테이크아웃',\n",
       " 870: '토마토',\n",
       " 871: '토스트',\n",
       " 872: '토요일',\n",
       " 873: '통닭',\n",
       " 874: '튀김',\n",
       " 875: '트리스',\n",
       " 876: '특색',\n",
       " 877: '특선',\n",
       " 878: '특유',\n",
       " 879: '특정',\n",
       " 880: '특징',\n",
       " 881: '티켓',\n",
       " 882: '파니니',\n",
       " 883: '파리',\n",
       " 884: '파스타',\n",
       " 885: '파이',\n",
       " 886: '파전',\n",
       " 887: '파크',\n",
       " 888: '파트너',\n",
       " 889: '파티',\n",
       " 890: '판매',\n",
       " 891: '팟타이',\n",
       " 892: '팥빙수',\n",
       " 893: '패키지',\n",
       " 894: '패티',\n",
       " 895: '팬케이크',\n",
       " 896: '편리',\n",
       " 897: '편안',\n",
       " 898: '평가',\n",
       " 899: '평균',\n",
       " 900: '평소',\n",
       " 901: '평양',\n",
       " 902: '평일',\n",
       " 903: '포인트',\n",
       " 904: '포장',\n",
       " 905: '포함',\n",
       " 906: '표시',\n",
       " 907: '표준',\n",
       " 908: '표현',\n",
       " 909: '푸드',\n",
       " 910: '품질',\n",
       " 911: '풍경',\n",
       " 912: '풍미',\n",
       " 913: '퓨전',\n",
       " 914: '프랑스',\n",
       " 915: '프랜차이즈',\n",
       " 916: '프레젠테이션',\n",
       " 917: '프렌치',\n",
       " 918: '프렌치 토스트',\n",
       " 919: '프로',\n",
       " 920: '프로모션',\n",
       " 921: '프리미엄',\n",
       " 922: '피자',\n",
       " 923: '필수',\n",
       " 924: '필요',\n",
       " 925: '하기',\n",
       " 926: '하나',\n",
       " 927: '하나하나',\n",
       " 928: '하네다',\n",
       " 929: '하녀',\n",
       " 930: '하루',\n",
       " 931: '하면',\n",
       " 932: '하얏트',\n",
       " 933: '하우스',\n",
       " 934: '하지',\n",
       " 935: '학생',\n",
       " 936: '한가지',\n",
       " 937: '한곳',\n",
       " 938: '한국',\n",
       " 939: '한국어',\n",
       " 940: '한국인',\n",
       " 941: '한번',\n",
       " 942: '한번쯤',\n",
       " 943: '한식',\n",
       " 944: '한옥',\n",
       " 945: '한우',\n",
       " 946: '한잔',\n",
       " 947: '한편',\n",
       " 948: '할랄',\n",
       " 949: '할인',\n",
       " 950: '합리',\n",
       " 951: '항목',\n",
       " 952: '해결',\n",
       " 953: '해당',\n",
       " 954: '해물',\n",
       " 955: '해변',\n",
       " 956: '해산물',\n",
       " 957: '해서',\n",
       " 958: '해외',\n",
       " 959: '해운대',\n",
       " 960: '해장',\n",
       " 961: '해장국',\n",
       " 962: '해주시',\n",
       " 963: '햄버거',\n",
       " 964: '행복',\n",
       " 965: '행사',\n",
       " 966: '향신료',\n",
       " 967: '허용',\n",
       " 968: '현금',\n",
       " 969: '현대',\n",
       " 970: '현지',\n",
       " 971: '현지인',\n",
       " 972: '호가',\n",
       " 973: '호떡',\n",
       " 974: '호박',\n",
       " 975: '호텔',\n",
       " 976: '혼자',\n",
       " 977: '혼잡',\n",
       " 978: '혼합',\n",
       " 979: '홍대',\n",
       " 980: '홍콩',\n",
       " 981: '홍합',\n",
       " 982: '화덕',\n",
       " 983: '화장실',\n",
       " 984: '확인',\n",
       " 985: '환경',\n",
       " 986: '환상',\n",
       " 987: '환영',\n",
       " 988: '황남빵',\n",
       " 989: '회사',\n",
       " 990: '회식',\n",
       " 991: '회의',\n",
       " 992: '회전',\n",
       " 993: '효율',\n",
       " 994: '후기',\n",
       " 995: '후식',\n",
       " 996: '후회',\n",
       " 997: '휴가',\n",
       " 998: '휴식',\n",
       " 999: '흥미'}"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "word_dict = dict(enumerate(words))\n",
    "word_dict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "from gensim.matutils import Sparse2Corpus\n",
    "corpus = Sparse2Corpus(tdm_cv.T)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(36477, 1000)"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tdm_cv.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "from gensim.models.ldamodel import LdaModel\n",
    "# 주제의 수는 사람이 결정\n",
    "# passes, iterations : 크면클수록 안정적 결과\n",
    "np.random.seed(123)\n",
    "lda = LdaModel(corpus=corpus, num_topics=100, passes=3,\n",
    "               iterations=100, id2word=word_dict)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('부산역', 0.32177186),\n",
       " ('요청', 0.22023596),\n",
       " ('타이', 0.12879394),\n",
       " ('발리', 0.12786213),\n",
       " ('위층', 0.08071742),\n",
       " ('위치', 0.03867282),\n",
       " ('음식', 0.029495724),\n",
       " ('주문', 0.014869005),\n",
       " ('식사', 0.009211435),\n",
       " ('고객', 0.0065540303)]"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lda.show_topic(0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('대부분', 0.25298467),\n",
       " ('관광', 0.11796317),\n",
       " ('남자', 0.081106946),\n",
       " ('브런치', 0.07762248),\n",
       " ('경치', 0.06614715),\n",
       " ('유리', 0.050314803),\n",
       " ('아이스', 0.04801052),\n",
       " ('해결', 0.047206227),\n",
       " ('실제', 0.0427407),\n",
       " ('음식', 0.030272555)]"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lda.show_topic(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(2, 1), (29, 1), (121, 1), (146, 1), (474, 1), (719, 2)]"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 문서의 주제 확인 : 단어 번호와 빈도로 이루어진 Bag-of-Words를 생성\n",
    "raw = tdm_cv[0]\n",
    "doc = list(zip(raw.indices, raw.data))\n",
    "doc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('가게', 1), ('감자탕', 1), ('김치', 1), ('느낌', 1), ('식당', 1), ('정도', 2)]"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "doc_words = [(words[i], n) for i, n in doc]\n",
    "doc_words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(16, 0.12625098),\n",
       " (37, 0.25585),\n",
       " (45, 0.17181301),\n",
       " (57, 0.14713661),\n",
       " (62, 0.180198)]"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lda.get_document_topics(doc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('가게', 0.259456),\n",
       " ('처음', 0.25258726),\n",
       " ('냉면', 0.11230256),\n",
       " ('지금', 0.102418154),\n",
       " ('자극', 0.041236725),\n",
       " ('생각', 0.037144694),\n",
       " ('방문', 0.03564293),\n",
       " ('프랜차이즈', 0.032939497),\n",
       " ('추천', 0.027422985),\n",
       " ('가격', 0.026906475)]"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lda.show_topic(62)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "lda.save('lda_test.lda')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "lda2 = LdaModel.load('lda_test.lda')"
   ]
  }
 ],
 "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}