시간대별 ppt 만들기용.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "aaeb5e2d",
   "metadata": {},
   "source": [
    "## 1. 리뷰내용 불러오기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "57fa5602",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import pickle\n",
    "from pprint import pprint\n",
    "import re"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "d505cd41",
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib # 글자 깨짐 방지 폰트설정\n",
    "from matplotlib import font_manager, rc\n",
    "import platform\n",
    "\n",
    "if platform.system() == 'Windows':\n",
    "# 윈도우인 경우\n",
    "    font_name = font_manager.FontProperties(fname=\"c:/Windows/Fonts/malgun.ttf\").get_name()\n",
    "    rc('font', family=font_name)\n",
    "else:    \n",
    "# Mac 인 경우\n",
    "    rc('font', family='AppleGothic')\n",
    "    \n",
    "matplotlib.rcParams['axes.unicode_minus'] = False   \n",
    "#그래프에서 마이너스 기호가 표시되도록 하는 설정입니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "30b15543",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "         DATE  STAR                                             REVIEW  LIKE\n",
      "0  2020-01-07     1          오류 자체 오류 이용 자수 결제 툭하면 점검 충전 체크카드 출금 충전 금액     0\n",
      "1  2020-01-07     5                          국굿 아이폰 바코드 큐알 코드 결제 가능 최고     0\n",
      "2  2020-01-07     3                           현상 사용 오늘 업데이트 접속 화면 바탕화면     0\n",
      "3  2020-01-07     2        설치 아이폰 설치 아버지 설치 갤럭시 노트 부산 지역화폐 설치 사용 버전 설치     0\n",
      "4  2020-01-07     4  개선 사항 요청 본인 인증 수단 공인 인증서 추가 어머니 발급 어머니 핸드폰 명의 ...     0\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 2266 entries, 0 to 2265\n",
      "Data columns (total 4 columns):\n",
      " #   Column  Non-Null Count  Dtype \n",
      "---  ------  --------------  ----- \n",
      " 0   DATE    2266 non-null   object\n",
      " 1   STAR    2266 non-null   int64 \n",
      " 2   REVIEW  2266 non-null   object\n",
      " 3   LIKE    2266 non-null   int64 \n",
      "dtypes: int64(2), object(2)\n",
      "memory usage: 70.9+ KB\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "with open(\"D:/cleaned_data.pk\", \"rb\") as f:\n",
    "    data = pickle.load(f)\n",
    "\n",
    "data.reset_index(drop=True, inplace=True)\n",
    "print(data.head())\n",
    "print(data.info())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8ff269d4",
   "metadata": {},
   "source": [
    "## 2. 분석시기 설정하기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "44f2992e",
   "metadata": {},
   "outputs": [],
   "source": [
    "##Date column을 pandas.Datetime으로 변환\n",
    "data['DATE'] = pd.to_datetime(data['DATE'])\n",
    "# Date column을 index로 설정\n",
    "data = data.set_index(\"DATE\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "670348c5",
   "metadata": {},
   "outputs": [],
   "source": [
    "## 4월을 기준으로 나누기\n",
    "before_4 = data['2019-12-27' : '2021-04-04']\n",
    "after_4 = data['2021-04-05': '2021-7-23']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "25def421",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "DatetimeIndex: 1333 entries, 2020-01-07 to 2021-04-04\n",
      "Data columns (total 3 columns):\n",
      " #   Column  Non-Null Count  Dtype \n",
      "---  ------  --------------  ----- \n",
      " 0   STAR    1333 non-null   int64 \n",
      " 1   REVIEW  1333 non-null   object\n",
      " 2   LIKE    1333 non-null   int64 \n",
      "dtypes: int64(2), object(1)\n",
      "memory usage: 41.7+ KB\n",
      "None\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "DatetimeIndex: 933 entries, 2021-04-05 to 2021-07-23\n",
      "Data columns (total 3 columns):\n",
      " #   Column  Non-Null Count  Dtype \n",
      "---  ------  --------------  ----- \n",
      " 0   STAR    933 non-null    int64 \n",
      " 1   REVIEW  933 non-null    object\n",
      " 2   LIKE    933 non-null    int64 \n",
      "dtypes: int64(2), object(1)\n",
      "memory usage: 29.2+ KB\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "# 확인해보기\n",
    "print(before_4.info())\n",
    "print(after_4.info())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0275f531",
   "metadata": {},
   "source": [
    "### 시기별로 나눠서 데이터 저장하기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "d82506c4",
   "metadata": {},
   "outputs": [],
   "source": [
    "# time_slice별로 데이터 갯수를 저장할 리스트 생성\n",
    "time_slice = []\n",
    "## 나눈 시점부터 리뷰 가져오기\n",
    "slice0 = list(before_4['REVIEW'])\n",
    "slice1 = list(after_4['REVIEW'])\n",
    "\n",
    "# LDA 분석할 때와 비슷하게 모든 분석 대상 텍스트를 담는 리스트 생성\n",
    "tokenized_data = [msg.split() for msg in (slice0 + slice1)]\n",
    "\n",
    "# 각 slice에 들어 있는 갯수를 원소로 갖는 리스트 생성\n",
    "time_slice.append(len(slice0))\n",
    "time_slice.append(len(slice1))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "7a18998c",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2266\n",
      "[1333, 933]\n",
      "['오류 자체 오류 이용 자수 결제 툭하면 점검 충전 체크카드 출금 충전 금액', '국굿 아이폰 바코드 큐알 코드 결제 가능 최고', '현상 사용 오늘 업데이트 접속 화면 바탕화면', '설치 아이폰 설치 아버지 설치 갤럭시 노트 부산 지역화폐 설치 사용 버전 설치', '개선 사항 요청 본인 인증 수단 공인 인증서 추가 어머니 발급 어머니 핸드폰 명의 사용 어머니 발급']\n",
      "['삼페 선불카드 선불카드', '그동안 사용 업체 지금 설치 인터페이스 이전 업체 사용성 이유 도대체', '다시 신규 카드 발급 지금 현재 카드 다시 카드 발급 사용 발급 신청 정상 설치 화면 인터페이스 이전 업체 이전 업체 인터페이스 화면 현찰 포인트 지금 화면 사용 가능 금액 표기 금액 포인트 포함 금액 사용 현찰 금액 표기 마지막 현재 카드 카드 사용 카드 현재 카드 사용 마지막 고객 센터 통화 계속 전화기 잡고 통화 인원 충당 방법', '사용자 편의 고려', '알바 달러 꼬름 선불카드 강매 전화통화 포인트 다시 운영']\n",
      "[['오류',\n",
      "  '자체',\n",
      "  '오류',\n",
      "  '이용',\n",
      "  '자수',\n",
      "  '결제',\n",
      "  '툭하면',\n",
      "  '점검',\n",
      "  '충전',\n",
      "  '체크카드',\n",
      "  '출금',\n",
      "  '충전',\n",
      "  '금액'],\n",
      " ['국굿', '아이폰', '바코드', '큐알', '코드', '결제', '가능', '최고'],\n",
      " ['현상', '사용', '오늘', '업데이트', '접속', '화면', '바탕화면'],\n",
      " ['설치',\n",
      "  '아이폰',\n",
      "  '설치',\n",
      "  '아버지',\n",
      "  '설치',\n",
      "  '갤럭시',\n",
      "  '노트',\n",
      "  '부산',\n",
      "  '지역화폐',\n",
      "  '설치',\n",
      "  '사용',\n",
      "  '버전',\n",
      "  '설치'],\n",
      " ['개선',\n",
      "  '사항',\n",
      "  '요청',\n",
      "  '본인',\n",
      "  '인증',\n",
      "  '수단',\n",
      "  '공인',\n",
      "  '인증서',\n",
      "  '추가',\n",
      "  '어머니',\n",
      "  '발급',\n",
      "  '어머니',\n",
      "  '핸드폰',\n",
      "  '명의',\n",
      "  '사용',\n",
      "  '어머니',\n",
      "  '발급']]\n"
     ]
    }
   ],
   "source": [
    "print(len(tokenized_data))\n",
    "print(time_slice)\n",
    "print(slice0[:5])\n",
    "print(slice1[:5])\n",
    "pprint(tokenized_data[:5])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5bdd7b5f",
   "metadata": {},
   "source": [
    "## 3.Dynamic Topic Model 돌리기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "9f7f5442",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\envs\\textmining\\lib\\site-packages\\gensim\\similarities\\__init__.py:15: UserWarning: The gensim.similarities.levenshtein submodule is disabled, because the optional Levenshtein package <https://pypi.org/project/python-Levenshtein/> is unavailable. Install Levenhstein (e.g. `pip install python-Levenshtein`) to suppress this warning.\n",
      "  warnings.warn(msg)\n"
     ]
    }
   ],
   "source": [
    "from gensim.models import ldaseqmodel\n",
    "from gensim.models import CoherenceModel\n",
    "from gensim.corpora import Dictionary, bleicorpus\n",
    "from gensim.matutils import hellinger\n",
    "from gensim import corpora\n",
    "from tqdm import tqdm_notebook\n",
    "from time import time\n",
    "\n",
    "import os"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "34696ecd",
   "metadata": {},
   "source": [
    "#### 문서 단어 행렬"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "71a4b5bd",
   "metadata": {},
   "outputs": [],
   "source": [
    "# OKT dictionary 저장.\n",
    "if not os.path.exists('dong(okt)_dict'):\n",
    "    dictionary = corpora.Dictionary(tokenized_data)\n",
    "    dictionary.save('dong(okt)_dict')\n",
    "    print(dictionary)\n",
    "else:\n",
    "    dictionary = Dictionary.load('dong(okt)_dict')\n",
    "\n",
    "# okt Corpus 저장.\n",
    "if not os.path.exists('dong(okt)_corpus'):\n",
    "    corpus = [dictionary.doc2bow(doc) for doc in tokenized_data]\n",
    "    corpora.BleiCorpus.serialize('dong(okt)_corpus', corpus)\n",
    "else:\n",
    "    corpus = bleicorpus.BleiCorpus('dong(okt)_corpus')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "73005fe6",
   "metadata": {},
   "source": [
    "### RUN DTM model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "780eb1d6",
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'corpus' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-11-d90121800b3b>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m      2\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      3\u001b[0m \u001b[1;32mif\u001b[0m \u001b[1;32mnot\u001b[0m \u001b[0mos\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mpath\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mexists\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m'last_dong(okt)_model'\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m----> 4\u001b[1;33m     dtm_model = ldaseqmodel.LdaSeqModel(corpus=corpus, id2word=dictionary,\n\u001b[0m\u001b[0;32m      5\u001b[0m                                         \u001b[0mtime_slice\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mtime_slice\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      6\u001b[0m                                         \u001b[0mnum_topics\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mNUM_TOPICS\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mNameError\u001b[0m: name 'corpus' is not defined"
     ]
    }
   ],
   "source": [
    "NUM_TOPICS = 8\n",
    "\n",
    "if not os.path.exists('last_dong(okt)_model'):\n",
    "    dtm_model = ldaseqmodel.LdaSeqModel(corpus=corpus, id2word=dictionary,\n",
    "                                        time_slice=time_slice, \n",
    "                                        num_topics=NUM_TOPICS,\n",
    "                                        passes=10,\n",
    "                                       random_state = 100,\n",
    "                                       alpha = \"auto\",\n",
    "                                       eta = \"auto\",\n",
    "                                       iterations=1000)\n",
    " \n",
    "    dtm_model.save('last_dong(okt)_model')\n",
    "else:\n",
    "    dtm_model = ldaseqmodel.LdaSeqModel.load('dong(okt)_model')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "baa1a45f",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "\n",
    "with open(\"dong(okt)_models.pk\", 'wb') as f:\n",
    "    pickle.dump(dtm_model, f)\n",
    "    \n",
    "# with open(\"dong(okt)_cs.pk\", 'wb') as f:\n",
    "#     pickle.dump(coherence_scores, f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "232175a1",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "\n",
    "with open(\"dong(okt)_models.pk\", 'rb') as f:\n",
    "    dtm_model = pickle.load(f)\n",
    "    \n",
    "# with open(\"dong_dtm_cs.pk\", 'rb') as f:\n",
    "#     coherence_scores = pickle.load(f)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c2436f17",
   "metadata": {},
   "source": [
    "## DTM 결과보기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "f6bad566",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[('최고', 0.07001649798931316),\n",
      "  ('포인트', 0.053602048229677804),\n",
      "  ('적립', 0.0521056152605506),\n",
      "  ('프로', 0.051559040845456),\n",
      "  ('답변', 0.04651368737304476),\n",
      "  ('다운', 0.03922151463627158),\n",
      "  ('회사', 0.026164036893195387),\n",
      "  ('현재', 0.025317303405731885),\n",
      "  ('에러', 0.023131949917973706),\n",
      "  ('문의', 0.022628526674195067),\n",
      "  ('사용', 0.021568832197757022),\n",
      "  ('예전', 0.020226310080671047),\n",
      "  ('버전', 0.019535381107736394),\n",
      "  ('가게', 0.01751299756545957),\n",
      "  ('지금', 0.015725736780723452),\n",
      "  ('이제', 0.015361431996847084),\n",
      "  ('감사', 0.01507301379432452),\n",
      "  ('대기', 0.014921422836049624),\n",
      "  ('세상', 0.014338640663679427),\n",
      "  ('대박', 0.013808324418750862)],\n",
      " [('가맹', 0.1120175720002429),\n",
      "  ('결제', 0.05867388394132755),\n",
      "  ('확인', 0.05196633086634966),\n",
      "  ('알림', 0.049918155758229346),\n",
      "  ('잔액', 0.043911515516504636),\n",
      "  ('찾기', 0.04219288161486457),\n",
      "  ('검색', 0.03786719217745438),\n",
      "  ('개선', 0.027582749040177743),\n",
      "  ('기능', 0.02436930662288557),\n",
      "  ('충전', 0.021857605917944976),\n",
      "  ('화면', 0.020607776059782514),\n",
      "  ('설정', 0.019379978487790023),\n",
      "  ('사용', 0.018217174999497248),\n",
      "  ('아이폰', 0.016232036226360342),\n",
      "  ('제발', 0.013444972947531721),\n",
      "  ('메뉴', 0.013084819570579423),\n",
      "  ('부분', 0.01239557153935567),\n",
      "  ('카드', 0.012104888312396356),\n",
      "  ('알람', 0.01196443393794761),\n",
      "  ('조회', 0.011160084691468869)],\n",
      " [('카드', 0.20679005713382204),\n",
      "  ('발급', 0.10091969349549905),\n",
      "  ('기존', 0.07430408385185028),\n",
      "  ('신청', 0.03800335146819772),\n",
      "  ('사용', 0.03499348775803182),\n",
      "  ('신규', 0.03319709237876355),\n",
      "  ('선불카드', 0.018840777331045205),\n",
      "  ('무조건', 0.015069930304065008),\n",
      "  ('취소', 0.01348482139952493),\n",
      "  ('연동', 0.011577968988921558),\n",
      "  ('재발', 0.011559972345986923),\n",
      "  ('강제', 0.011364212036250461),\n",
      "  ('등록', 0.010631340097722677),\n",
      "  ('부산시', 0.008918834378543884),\n",
      "  ('시스템', 0.008408890984670377),\n",
      "  ('계속', 0.008327022111266241),\n",
      "  ('설명', 0.00807703151445029),\n",
      "  ('낭비', 0.007965620544842603),\n",
      "  ('쓰레기', 0.007889877337175319),\n",
      "  ('안내', 0.007653317407292289)],\n",
      " [('카드', 0.08692670102539318),\n",
      "  ('페이', 0.056453715230779994),\n",
      "  ('등록', 0.05280328611593053),\n",
      "  ('체크카드', 0.03393526484134785),\n",
      "  ('혜택', 0.03362162280999482),\n",
      "  ('사용', 0.032335924263596265),\n",
      "  ('삼성', 0.029195332988704043),\n",
      "  ('실물', 0.020671468543909687),\n",
      "  ('선불카드', 0.02003989547339197),\n",
      "  ('기존', 0.01565115484846714),\n",
      "  ('그냥', 0.01470654663903581),\n",
      "  ('결재', 0.014110988493151053),\n",
      "  ('이벤트', 0.010861265879871988),\n",
      "  ('발급', 0.010813560017732371),\n",
      "  ('소득공제', 0.010747348594057038),\n",
      "  ('교통카드', 0.010123567903492958),\n",
      "  ('입력', 0.009741171089298293),\n",
      "  ('동의', 0.009557612701443032),\n",
      "  ('정보', 0.00938395162493835),\n",
      "  ('신청', 0.009055884611001616)],\n",
      " [('오류', 0.04667621678231401),\n",
      "  ('접속', 0.04157349845065354),\n",
      "  ('인증', 0.034542669403132414),\n",
      "  ('실행', 0.032079970223627546),\n",
      "  ('연결', 0.031507987730594045),\n",
      "  ('네트워크', 0.027582813625545977),\n",
      "  ('충전', 0.02681113246669785),\n",
      "  ('다시', 0.02559432180647407),\n",
      "  ('계좌', 0.02460947539459961),\n",
      "  ('계속', 0.02431342424938992),\n",
      "  ('번호', 0.020089881776700815),\n",
      "  ('고객', 0.01810339587290415),\n",
      "  ('확인', 0.017104027442034517),\n",
      "  ('센터', 0.017065870255973132),\n",
      "  ('문제', 0.01459654926550188),\n",
      "  ('와이파이', 0.013965896081336705),\n",
      "  ('사용', 0.01356597437881003),\n",
      "  ('데이터', 0.012045465767445645),\n",
      "  ('상태', 0.011825287092873029),\n",
      "  ('은행', 0.011113707175401905)],\n",
      " [('결제', 0.07940334427157032),\n",
      "  ('카드', 0.05940968704380691),\n",
      "  ('기능', 0.03116081733419574),\n",
      "  ('충전', 0.024874249686484856),\n",
      "  ('부산', 0.021688204603877),\n",
      "  ('사용자', 0.018405422472329052),\n",
      "  ('지역', 0.017674056163574377),\n",
      "  ('서비스', 0.015792047406151186),\n",
      "  ('업체', 0.015588922566293338),\n",
      "  ('경제', 0.013500419773975468),\n",
      "  ('한도', 0.01308103847873084),\n",
      "  ('바코드', 0.0125279372390363),\n",
      "  ('운영', 0.012362804186594177),\n",
      "  ('이전', 0.011306576834502823),\n",
      "  ('부산시', 0.010095764353523232),\n",
      "  ('저번', 0.00816882642605596),\n",
      "  ('편의', 0.007996688030558786),\n",
      "  ('코드', 0.007888937833961763),\n",
      "  ('미리', 0.007840228673478958),\n",
      "  ('기존', 0.007412697677677996)],\n",
      " [('불편', 0.04397140548228035),\n",
      "  ('최악', 0.04237738566236265),\n",
      "  ('할인', 0.03088028734346712),\n",
      "  ('이전', 0.030070539595987482),\n",
      "  ('리뷰', 0.028769191537835574),\n",
      "  ('도움', 0.027942107073464912),\n",
      "  ('사람', 0.026750527611176998),\n",
      "  ('사용', 0.026719229637009942),\n",
      "  ('사항', 0.02321888442391216),\n",
      "  ('하루', 0.02155839172768698),\n",
      "  ('생각', 0.021282508527561),\n",
      "  ('직관', 0.0203343478906113),\n",
      "  ('종일', 0.01832631487784268),\n",
      "  ('디자인', 0.016947473679397695),\n",
      "  ('자주', 0.01683594122628338),\n",
      "  ('이름', 0.015064707799011616),\n",
      "  ('불만', 0.015008295883891463),\n",
      "  ('이해', 0.014131732156419727),\n",
      "  ('일부러', 0.012763363306875617),\n",
      "  ('정신', 0.01183670733143384)],\n",
      " [('사용', 0.18837709253817297),\n",
      "  ('금액', 0.08185845471569131),\n",
      "  ('캐시', 0.07722223187620184),\n",
      "  ('충전', 0.0575438006272783),\n",
      "  ('확인', 0.01994907219644027),\n",
      "  ('이전', 0.019348864841330478),\n",
      "  ('이용', 0.017541536853402134),\n",
      "  ('잔액', 0.017036197985722146),\n",
      "  ('환불', 0.016253070276860827),\n",
      "  ('포인트', 0.01517355267576183),\n",
      "  ('캐쉬백', 0.014270245257732777),\n",
      "  ('적립', 0.014208904331506171),\n",
      "  ('가능', 0.012794060292402511),\n",
      "  ('표시', 0.010667038607852756),\n",
      "  ('개선', 0.010491687250259969),\n",
      "  ('예전', 0.009789928809971923),\n",
      "  ('합계', 0.009658905865894294),\n",
      "  ('설정', 0.009155781874257255),\n",
      "  ('혜택', 0.008962298165462864),\n",
      "  ('결제', 0.008491140203109312)],\n",
      " [('업데이트', 0.14846033539707354),\n",
      "  ('부산', 0.09422825543406166),\n",
      "  ('설치', 0.08900573076245621),\n",
      "  ('시민', 0.03526939392476994),\n",
      "  ('다시', 0.0339925519614993),\n",
      "  ('사용', 0.02450203154445032),\n",
      "  ('삭제', 0.023856390086579523),\n",
      "  ('계속', 0.020335172876408944),\n",
      "  ('쓰레기', 0.020054375784526422),\n",
      "  ('운영', 0.017575950291083182),\n",
      "  ('기대', 0.015789944337829858),\n",
      "  ('지역화폐', 0.01557277200974277),\n",
      "  ('업뎃', 0.012793210102476416),\n",
      "  ('시청', 0.012464357514558875),\n",
      "  ('짜증', 0.012233443600806505),\n",
      "  ('개선', 0.011614841149157574),\n",
      "  ('문제', 0.011433878525475626),\n",
      "  ('수준', 0.010983706810365767),\n",
      "  ('화면', 0.010871194860376493),\n",
      "  ('선정', 0.009769163110246218)]]\n"
     ]
    }
   ],
   "source": [
    "# 고정된 시간 내에서 전체 토픽보기\n",
    "pprint(dtm_model.print_topics(time=1, top_terms=20))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "23a3fdb5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[[('최고', 0.07062006250042156),\n",
       "  ('포인트', 0.053411398187502734),\n",
       "  ('적립', 0.05200975139860206),\n",
       "  ('프로', 0.05162119458100169),\n",
       "  ('답변', 0.04570841865409775),\n",
       "  ('다운', 0.04057125755208223),\n",
       "  ('회사', 0.025873993886910654),\n",
       "  ('현재', 0.025282645222782948),\n",
       "  ('에러', 0.02320360521152509),\n",
       "  ('문의', 0.022523918860693073),\n",
       "  ('사용', 0.0215423669056965),\n",
       "  ('예전', 0.01984606429409958),\n",
       "  ('버전', 0.019496561255528927),\n",
       "  ('가게', 0.017637146243612914),\n",
       "  ('지금', 0.015636670151540997),\n",
       "  ('이제', 0.015298179487032446),\n",
       "  ('감사', 0.015063061645856742),\n",
       "  ('대기', 0.014875866807749247),\n",
       "  ('세상', 0.014389541174000113),\n",
       "  ('대박', 0.01387743004762505)],\n",
       " [('최고', 0.07001649798931316),\n",
       "  ('포인트', 0.053602048229677804),\n",
       "  ('적립', 0.0521056152605506),\n",
       "  ('프로', 0.051559040845456),\n",
       "  ('답변', 0.04651368737304476),\n",
       "  ('다운', 0.03922151463627158),\n",
       "  ('회사', 0.026164036893195387),\n",
       "  ('현재', 0.025317303405731885),\n",
       "  ('에러', 0.023131949917973706),\n",
       "  ('문의', 0.022628526674195067),\n",
       "  ('사용', 0.021568832197757022),\n",
       "  ('예전', 0.020226310080671047),\n",
       "  ('버전', 0.019535381107736394),\n",
       "  ('가게', 0.01751299756545957),\n",
       "  ('지금', 0.015725736780723452),\n",
       "  ('이제', 0.015361431996847084),\n",
       "  ('감사', 0.01507301379432452),\n",
       "  ('대기', 0.014921422836049624),\n",
       "  ('세상', 0.014338640663679427),\n",
       "  ('대박', 0.013808324418750862)]]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 고정된 토픽 내에서 전체 시간 변화 보기\n",
    "dtm_model.print_topic_times(topic=0, top_terms=20)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b9883f18",
   "metadata": {},
   "source": [
    "## 6. 문서별 토픽 비중 구하기\n",
    "> 1) 토픽 비중 계산을 위한 데이터 전처리\n",
    "\n",
    "> 2) OOM 문제 해결을 위한 샘플링\n",
    "\n",
    "> 3) DataFrame으로 만들어 excel 파일로 저장"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "8458ea58",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2266\n"
     ]
    }
   ],
   "source": [
    "doc_time = []\n",
    "\n",
    "for doc_id in range(len(tokenized_data)):\n",
    "    if doc_id < time_slice[0]:\n",
    "        doc_time.append(0)\n",
    "    else:\n",
    "        doc_time.append(1)\n",
    "    \n",
    "        \n",
    "print(len(doc_time))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "95ce9de5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 952  113  939 1235 1047  865   42  108  916 1191]\n",
      "[1597 1396 1338 1856 1355 2255 1406 1806 2194 1390]\n"
     ]
    }
   ],
   "source": [
    "# time_slot 별로 900개씩만 뽑기.\n",
    "sample_idx1 = np.random.randint(0, time_slice[0], size=900)\n",
    "sample_idx2 = np.random.randint(time_slice[0], time_slice[0] + time_slice[1],\n",
    "                                size=900)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "6a5fe9ee",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1800\n",
      "1800\n"
     ]
    }
   ],
   "source": [
    "# OOM(Out-of-Memory) 문제로 인해 1000개씩 샘플링 하여 실행.\n",
    "\n",
    "processing_docs = []\n",
    "processing_time = []\n",
    "\n",
    "sample_indices = np.concatenate((sample_idx1, sample_idx2))\n",
    "\n",
    "for idx in sample_indices:\n",
    "    processing_docs.append(tokenized_data[idx])\n",
    "    processing_time.append(doc_time[idx])\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "2d835d17",
   "metadata": {},
   "outputs": [],
   "source": [
    "doc_dist = []\n",
    "\n",
    "for doc_id in sample_indices:\n",
    "    doc_dist.append(dtm_model.doc_topics(doc_id))\n",
    "    doc_dist = np.array(doc_dist)\n",
    "    \n",
    "doc_topic_dist = doc_dist.T\n",
    "doc_topic_dist.shape\n",
    "\n",
    "NUM_TOPICS = dtm_model.num_topics\n",
    "\n",
    "for i in range(NUM_TOPICS):\n",
    "    print(\"%f\" % doc_dist[0][i], end=\", \")\n",
    "\n",
    "print()\n",
    "\n",
    "for i in range(NUM_TOPICS):\n",
    "    print(\"%f\" % doc_topic_dist[i][0], end=\", \")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "92adbdec",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1800,)"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "doc_topic_dist[0].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "be9b0f32",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   Time    Topic0    Topic1    Topic2    Topic3    Topic4    Topic5    Topic6  \\\n",
      "0     0  0.001642  0.001642  0.001642  0.001642  0.001642  0.001642  0.581646   \n",
      "1     0  0.001100  0.526072  0.001100  0.110039  0.087428  0.001100  0.001100   \n",
      "2     0  0.004785  0.004785  0.004785  0.004785  0.004785  0.004785  0.004785   \n",
      "3     0  0.004785  0.004785  0.004785  0.004785  0.482376  0.004785  0.004785   \n",
      "4     0  0.000827  0.000827  0.000827  0.000827  0.162540  0.000827  0.000827   \n",
      "\n",
      "     Topic7    Topic8  \n",
      "0  0.406859  0.001642  \n",
      "1  0.270959  0.001100  \n",
      "2  0.961722  0.004785  \n",
      "3  0.004785  0.484131  \n",
      "4  0.421942  0.410556  \n"
     ]
    }
   ],
   "source": [
    "dtm_df = pd.DataFrame({\"Time\" : processing_time, \"Topic1\" : doc_topic_dist[0], \"Topic2\" : doc_topic_dist[1],\n",
    "                      \"Topic2\" : doc_topic_dist[2], \"Topic3\" : doc_topic_dist[3], \"Topic4\" : doc_topic_dist[4]\n",
    "                    , \"Topic5\" : doc_topic_dist[5], \"Topic6\" : doc_topic_dist[6], \"Topic7\" : doc_topic_dist[7],\n",
    "                       \"Topic8\" : doc_topic_dist[8]\n",
    "                      })\n",
    "\n",
    "print(dtm_df.head())\n",
    "dtm_df.to_excel(\"dong100_dtm_12topic_dist.xlsx\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2eaa0d2b",
   "metadata": {},
   "source": [
    "## 7. 시간대별 토픽비중 변화 추이\n",
    "> 1) 토픽 비중 변화를 보기 위한 데이터 전처리\n",
    "\n",
    "> 2) 시간대별 토픽 비중 평균 계산"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "ca814b5b",
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'dtm_df' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-1-bf5c64b00f58>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m      1\u001b[0m \u001b[1;31m# groupby 오브젝트 생성하기\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m----> 2\u001b[1;33m \u001b[0mdoctopic_timeslot\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mdtm_df\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mgroupby\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m'Time'\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      3\u001b[0m \u001b[1;31m# doctopic_timeslot.groups\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      4\u001b[0m \u001b[1;31m# mean() 함수 이용해서 Topic0~TopicN까지의 모든 열 평균 구하기\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      5\u001b[0m \u001b[0mtimeslot_avg\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mdoctopic_timeslot\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mmean\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mNameError\u001b[0m: name 'dtm_df' is not defined"
     ]
    }
   ],
   "source": [
    "# groupby 오브젝트 생성하기\n",
    "doctopic_timeslot = dtm_df.groupby('Time')\n",
    "doctopic_timeslot.groups\n",
    "# mean() 함수 이용해서 Topic0~TopicN까지의 모든 열 평균 구하기\n",
    "timeslot_avg = doctopic_timeslot.mean()\n",
    "timeslot_avg = timeslot_avg.reset_index()   # index 리셋하기\n",
    "timeslot_avg"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "ae48ff74",
   "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>Time</th>\n",
       "      <th>Topic0</th>\n",
       "      <th>Topic1</th>\n",
       "      <th>Topic2</th>\n",
       "      <th>Topic3</th>\n",
       "      <th>Topic4</th>\n",
       "      <th>Topic5</th>\n",
       "      <th>Topic6</th>\n",
       "      <th>Topic7</th>\n",
       "      <th>Topic8</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0.099541</td>\n",
       "      <td>0.067436</td>\n",
       "      <td>0.062597</td>\n",
       "      <td>0.050793</td>\n",
       "      <td>0.271521</td>\n",
       "      <td>0.096385</td>\n",
       "      <td>0.056102</td>\n",
       "      <td>0.183136</td>\n",
       "      <td>0.112490</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0.060739</td>\n",
       "      <td>0.126155</td>\n",
       "      <td>0.205124</td>\n",
       "      <td>0.131779</td>\n",
       "      <td>0.072033</td>\n",
       "      <td>0.104924</td>\n",
       "      <td>0.069071</td>\n",
       "      <td>0.152031</td>\n",
       "      <td>0.078145</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Time    Topic0    Topic1    Topic2    Topic3    Topic4    Topic5    Topic6  \\\n",
       "0     0  0.099541  0.067436  0.062597  0.050793  0.271521  0.096385  0.056102   \n",
       "1     1  0.060739  0.126155  0.205124  0.131779  0.072033  0.104924  0.069071   \n",
       "\n",
       "     Topic7    Topic8  \n",
       "0  0.183136  0.112490  \n",
       "1  0.152031  0.078145  "
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# mean() 함수 이용해서 Topic0~TopicN까지의 모든 열 평균 구하기\n",
    "timeslot_avg = doctopic_timeslot.mean()\n",
    "timeslot_avg = timeslot_avg.reset_index()   # index 리셋하기\n",
    "timeslot_avg"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "6d4d7e03",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    1.0\n",
       "1    2.0\n",
       "dtype: float64"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 시기별 모든 토픽 값들의 합은 1이다.\n",
    "timeslot_avg.sum(axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "05d4cc40",
   "metadata": {},
   "source": [
    "## 8. 6번7번 결과를 그래프로"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "id": "b8c3442f",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x864 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# figure 하나당 4개씩 그래프 그리기\n",
    "from matplotlib import pyplot as plt\n",
    "plt.figure(figsize=(16, 12)) \n",
    "plt.plot(timeslot_avg['Time'], timeslot_avg['Topic0'], 'b*--', label='토픽0')\n",
    "plt.plot(timeslot_avg['Time'], timeslot_avg['Topic1'], 'rs--', label='토픽1')\n",
    "plt.plot(timeslot_avg['Time'], timeslot_avg['Topic2'], 'g^--', label='토픽2')\n",
    "plt.plot(timeslot_avg['Time'], timeslot_avg['Topic3'], 'y*--', label='토픽3')\n",
    "plt.plot(timeslot_avg['Time'], timeslot_avg['Topic4'], 'bs--', label='토픽4')\n",
    "plt.plot(timeslot_avg['Time'], timeslot_avg['Topic5'], 'm*--', label='토픽5')\n",
    "plt.plot(timeslot_avg['Time'], timeslot_avg['Topic6'], 'k*--', label='토픽6')\n",
    "plt.plot(timeslot_avg['Time'], timeslot_avg['Topic7'], 'c*--', label='토픽7')\n",
    "plt.plot(timeslot_avg['Time'], timeslot_avg['Topic8'], 'b--', label='토픽8')\n",
    "\n",
    "#그래프에서 값 표시하는 annotate 방법 찾기\n",
    "for topic_id in ['Topic0', 'Topic1', 'Topic2', 'Topic3', 'Topic4', 'Topic5', \n",
    "                 'Topic6', 'Topic7', 'Topic8']:\n",
    "    for x, y in zip(range(len(time_slice)), timeslot_avg[topic_id]):\n",
    "        plt.annotate(\"%.4f\"%y, (x, y), textcoords=\"offset points\", \n",
    "                     xytext=(0,10), ha='center')\n",
    "\n",
    "plt.title(\"12개의 대화 토픽\", fontsize=14)\n",
    "plt.xlabel(\"대화 시기\", fontsize=13)\n",
    "plt.ylabel(\"토픽의 비중 (%)\", fontsize=13)\n",
    "plt.xticks(timeslot_avg['Time'], ['4월 이전','4월 이후'])\n",
    "plt.ylim([0.00, 0.3])\n",
    "plt.legend(loc=\"best\")\n",
    "plt.savefig(\"dong100_topic_distribution_graph.png\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "4dd23189",
   "metadata": {},
   "outputs": [],
   "source": [
    "from collections import defaultdict\n",
    "# 단어의 확률 변화를 보고 싶은 topic 번호 \n",
    "topic_id = 4\n",
    "# 해당 토픽에서 포함할 단어목록 결정\n",
    "topic_info = dtm_model.print_topic_times(topic=topic_id, top_terms=10)\n",
    "topic_dict = defaultdict(list)\n",
    "\n",
    "#time별 특정 topic(topic_id)의 단어별 확률값.\n",
    "for dist_inTime in topic_info:\n",
    "    #time별로 단어와 확률값 저장.\n",
    "    for word, prob in dist_inTime:\n",
    "        topic_dict[word].append(prob)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "52f44101",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "defaultdict(<class 'list'>, {'오류': [0.04671879723063112, 0.04667621678231401], '접속': [0.041795583798141946, 0.04157349845065354], '인증': [0.0346345163586488, 0.034542669403132414], '실행': [0.03221799222335726, 0.032079970223627546], '연결': [0.030603231897875824, 0.031507987730594045], '네트워크': [0.02774968606663024, 0.027582813625545977], '충전': [0.026832876228462244, 0.02681113246669785], '다시': [0.02564248801880364, 0.02559432180647407], '계좌': [0.024370127794495523, 0.02460947539459961], '계속': [0.024354228040320957, 0.02431342424938992]})\n"
     ]
    }
   ],
   "source": [
    "print(topic_dict)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "id": "313dd19b",
   "metadata": {},
   "outputs": [],
   "source": [
    "from collections import defaultdict\n",
    "# 단어의 확률 변화를 보고 싶은 topic 번호 \n",
    "topic_id = 2\n",
    "# 해당 토픽에서 포함할 단어목록 결정\n",
    "topic_info = dtm_model.print_topic_times(topic=topic_id, top_terms=10)\n",
    "topic_dict = defaultdict(list)\n",
    "\n",
    "#time별 특정 topic(topic_id)의 단어별 확률값.\n",
    "for dist_inTime in topic_info:\n",
    "    #time별로 단어와 확률값 저장.\n",
    "    for word, prob in dist_inTime:\n",
    "        topic_dict[word].append(prob)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "id": "e3c70caf",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "defaultdict(<class 'list'>, {'카드': [0.20755580956045394, 0.20679005713382204], '발급': [0.10116974612705028, 0.10091969349549905], '기존': [0.06981979629428169, 0.07430408385185028], '신청': [0.03826362427688667, 0.03800335146819772], '사용': [0.03498957824698463, 0.03499348775803182], '신규': [0.03308866621720359, 0.03319709237876355], '선불카드': [0.018760258850626172, 0.018840777331045205], '무조건': [0.0151459072015973, 0.015069930304065008], '취소': [0.013539268839719388, 0.01348482139952493], '재발': [0.011546031199129276], '연동': [0.011577968988921558]})\n"
     ]
    }
   ],
   "source": [
    "print(topic_dict)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "id": "2b0a7310",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x864 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#확률값 추이를 보고싶은 단어 목록\n",
    "keywords = [\"오류\",\"접속\",\"네트워크\",\"인증\"]\n",
    "\n",
    "# time_slot\n",
    "t = [i for i in range(len(time_slice))]\n",
    "\n",
    "plt.figure(figsize=(16, 12))\n",
    "for keyword in keywords:\n",
    "    plt.plot(t, topic_dict[keyword], label=keyword)\n",
    "    \n",
    "    for x, y in zip(t, topic_dict[keyword]):\n",
    "         plt.annotate(\"%.4f\"%y, (x, y), textcoords=\"offset points\", xytext=(0,10), ha='center')\n",
    "\n",
    "plt.title(\"시간별 단어의 토픽비중 변화\")\n",
    "plt.xlabel(\"시간\")\n",
    "plt.ylabel(\"토픽 비중\")\n",
    "plt.legend(loc='best')\n",
    "plt.savefig(\"kakao_keywords_distribution_overtime_graph.png\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f54e94bc",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0f934b6f",
   "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": "3.6.10"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": false,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}