korea_nltk.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('품사', 'NNG'),\n",
       " ('태', 'NNG'),\n",
       " ('깅을', 'UNKNOWN'),\n",
       " ('지원', 'NNG'),\n",
       " ('합니다', 'XSV+EC')]"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from eunjeon import Mecab # 은전한닢 프로젝트 한국어 형태소 분석기\n",
    "mecab = Mecab(dicpath='C:/mecab/mecab-ko-dic')  \n",
    "mecab.pos(\"품사 태깅을 지원합니다\") "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['형태소', '분리', '를', '지원', '합니다']"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mecab.morphs(\"형태소 분리를 지원합니다\") "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['명사', '해당', '형태소', '추출']"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mecab.nouns(\"명사에 해당하는 형태소만 추출합니다\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['네, 안녕하세요.', '반갑습니다.']\n"
     ]
    }
   ],
   "source": [
    "# https://konlpy-ko.readthedocs.io/ko/v0.4.3/ \n",
    "# 한국어 정보처리\n",
    "from konlpy.tag import Kkma\n",
    "from konlpy.utils import pprint\n",
    "kkma = Kkma()\n",
    "pprint(kkma.sentences(u'네, 안녕하세요. 반갑습니다.'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "안녕하세요ㅋㅋㅋ 반가워요 사랑해\n",
      "[('안녕하세요', 'Adjective'), ('.', 'Punctuation'), ('오래간만', 'Adverb'), ('이네요', 'Verb'), ('~~.', 'Punctuation'), ('어제', 'Noun'), ('재미있었어요', 'Adjective'), ('.', 'Punctuation')]\n",
      "['장', '변동', '폭', '보', '코스피', '사흘', '폭', '반등']\n"
     ]
    }
   ],
   "source": [
    "from konlpy.tag import Okt\n",
    "from collections import Counter\n",
    "# Okt 객체 생성\n",
    "okt = Okt()\n",
    "text = '안녕하세욬ㅋㅋㅋ 반가워요 샤릉해'\n",
    "print(okt.normalize(text))\n",
    "text = '안녕하세요. 오래간만이네요~~. 어제 재미있었어요.' \n",
    "print(okt.pos(text)) # 품사\n",
    "text = \"장중 큰 변동 폭을 보이던 코스피가 사흘 만에 소폭 반등했다.\" \n",
    "print(okt.nouns(text)) # 명사 추출\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['이번', '개봉', '영화', '배우', '연기력', '목소리']"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 불용어 제거 명사만 추출 \n",
    "from konlpy.tag import Mecab\n",
    "mecab = Mecab(dicpath='C:/mecab/mecab-ko-dic') \n",
    "\n",
    "stopwords = ['의','이','있','하','들','그','되','수','보','않','없','나','사람','아','등','같','오','있','한'] #불용어사전\n",
    "\n",
    "ex_text = \"이번에 새롭게 개봉한 영화의 배우들은 모두 훌륭한 연기력과 아름다운 목소리를 갖고 있어!!\"\n",
    "[i for i in mecab.nouns(ex_text) if i not in stopwords]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['이번', '에', '새롭게', '개봉', '한', '영화', '의', '배우', '들', '모두', '훌륭한', '연기력', '과', '아름다운', '목소리', '갖고', '있어']\n"
     ]
    }
   ],
   "source": [
    "# 불용어 제거\n",
    "import re\n",
    "from konlpy.tag import Okt\n",
    "tokenizer = Okt()\n",
    "def text_preprocessing(text,tokenizer):\n",
    "    \n",
    "    stopwords = ['을', '를', '이', '가', '은', '는']\n",
    "    \n",
    "    txt = re.sub('[^가-힣a-z]', ' ', text)\n",
    "    token = tokenizer.morphs(txt)\n",
    "    token = [t for t in token if t not in stopwords]\n",
    "        \n",
    "    return token\n",
    "\n",
    "ex_text = \"이번에 새롭게 개봉한 영화의 배우들은 모두 훌륭한 연기력과 아름다운 목소리를 갖고 있어!!\"\n",
    "example_pre= text_preprocessing(ex_text,tokenizer)\n",
    "print(example_pre)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "이번에 새롭게 개봉한 영화의 배우들은  훌륭한 연기력과 아름다운 목소리를 갖고 있어 \n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "\"\\ndef extract_word(text):\\n    hangul = re.compile('[^가-힣]') \\n    result = hangul.sub(' ', text) \\n    return result\\n\""
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 정규식 한글만 남기기\n",
    "ex_text = \"이번에 새롭게 개봉한 영화의 배우들은 everybody 훌륭한 연기력과 아름다운 목소리를 갖고 있어!! wow11!!\"\n",
    "review_text=re.sub(\"[^ㄱ-ㅎㅏ-ㅣ가-힣 ]\",\"\",ex_text)\n",
    "print(review_text)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['사과', '바나나']\n"
     ]
    }
   ],
   "source": [
    "# 한글자 제거\n",
    "words = ['귤','사과','바나나']\n",
    "\n",
    "words = [x for x in words if len(x)>1]\n",
    "print(words)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting git+https://github.com/ssut/py-hanspell.git\n",
      "  Cloning https://github.com/ssut/py-hanspell.git to c:\\users\\edu007\\appdata\\local\\temp\\pip-req-build-fbsrdo9s\n",
      "  Resolved https://github.com/ssut/py-hanspell.git to commit 8e993cf46f97f9d665c15633a0fc78ac1b727713\n",
      "  Preparing metadata (setup.py): started\n",
      "  Preparing metadata (setup.py): finished with status 'done'\n",
      "Requirement already satisfied: requests in c:\\users\\edu007\\.conda\\envs\\project\\lib\\site-packages (from py-hanspell==1.1) (2.26.0)\n",
      "Requirement already satisfied: urllib3<1.27,>=1.21.1 in c:\\users\\edu007\\.conda\\envs\\project\\lib\\site-packages (from requests->py-hanspell==1.1) (1.26.7)\n",
      "Requirement already satisfied: charset-normalizer~=2.0.0 in c:\\users\\edu007\\.conda\\envs\\project\\lib\\site-packages (from requests->py-hanspell==1.1) (2.0.7)\n",
      "Requirement already satisfied: idna<4,>=2.5 in c:\\users\\edu007\\.conda\\envs\\project\\lib\\site-packages (from requests->py-hanspell==1.1) (3.2)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in c:\\users\\edu007\\.conda\\envs\\project\\lib\\site-packages (from requests->py-hanspell==1.1) (2021.5.30)\n",
      "Building wheels for collected packages: py-hanspell\n",
      "  Building wheel for py-hanspell (setup.py): started\n",
      "  Building wheel for py-hanspell (setup.py): finished with status 'done'\n",
      "  Created wheel for py-hanspell: filename=py_hanspell-1.1-py3-none-any.whl size=4906 sha256=611740f04b2adc14bbac210b880aa80f9afdb91ba782a5f12645410acaa89fa0\n",
      "  Stored in directory: C:\\Users\\edu007\\AppData\\Local\\Temp\\pip-ephem-wheel-cache-d3d8s6pl\\wheels\\ab\\f5\\7b\\d4124bb329c905301baed80e2ae45aa14e824f62ebc3ec2cc4\n",
      "Successfully built py-hanspell\n",
      "Installing collected packages: py-hanspell\n",
      "Successfully installed py-hanspell-1.1\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  Running command git clone --filter=blob:none -q https://github.com/ssut/py-hanspell.git 'C:\\Users\\edu007\\AppData\\Local\\Temp\\pip-req-build-fbsrdo9s'\n"
     ]
    }
   ],
   "source": [
    "!pip install git+https://github.com/ssut/py-hanspell.git"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "맞춤법이 틀린 문장이 되었습니다.\n"
     ]
    }
   ],
   "source": [
    "# 맞춤법 교정\n",
    "from hanspell import spell_checker\n",
    "\n",
    "sent = \"맞춤법이 틀린 문장이 돼었습니다.\"\n",
    "spelled_sent = spell_checker.check(sent)\n",
    "\n",
    "hanspell_sent = spelled_sent.checked\n",
    "print(hanspell_sent)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "showing info https://raw.githubusercontent.com/nltk/nltk_data/gh-pages/index.xml\n"
     ]
    }
   ],
   "source": [
    "import nltk\n",
    "nltk.download()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['네, 안녕하세요.', '반갑습니다.']\n"
     ]
    }
   ],
   "source": [
    "# 문단 -> 문장 리스트 형식\n",
    "from konlpy.tag import Kkma\n",
    "from konlpy.utils import pprint\n",
    "kkma = Kkma()\n",
    "pprint(kkma.sentences(u'네, 안녕하세요. 반갑습니다.'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['자연어 처리 하고 있습니다.', '하바나온난화 나나나나 합니다.', '오늘 점심은 뭐죠?']\n"
     ]
    }
   ],
   "source": [
    "# 문장 단위 자르기\n",
    "from nltk.tokenize import sent_tokenize\n",
    "text = '자연어 처리 하고 있습니다. 하바나온난화 나나나나 합니다. 오늘 점심은 뭐죠?'\n",
    "print(sent_tokenize(text))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package punkt to C:\\nltk_data...\n",
      "[nltk_data]   Unzipping tokenizers\\punkt.zip.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import nltk\n",
    "nltk.download('punkt')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('우리', 5)\n",
      "('봄바람', 5)\n",
      "('얼음', 4)\n",
      "('가치', 4)\n",
      "('이상', 4)\n",
      "('얼마나', 4)\n",
      "('것', 3)\n",
      "('황금시대', 3)\n",
      "('작고', 3)\n",
      "('긴지', 3)\n"
     ]
    }
   ],
   "source": [
    "# 명사 빈도 계산\n",
    "from konlpy.tag import Okt\n",
    "from collections import Counter\n",
    "\n",
    "f = open('test.txt', 'r', encoding='utf-8')\n",
    "text = f.read()\n",
    "\n",
    "# okt 객체 생성\n",
    "okt = Okt()\n",
    "noun = okt.nouns(text)\n",
    "# 한글 자 명사 제외\n",
    "for i,v in enumerate(noun):\n",
    "    if len(v) < 2 :\n",
    "        noun.pop(i)\n",
    "\n",
    "count = Counter(noun)\n",
    "\n",
    "# 명사 빈도 카운트\n",
    "noun_list = count.most_common(10)\n",
    "for v in noun_list:\n",
    "    print(v)\n",
    "\n",
    "# 파일 저장\n",
    "with open('noun_list.csv','w',newline='',encoding='euc-kr') as f:\n",
    "    csvw = csv.writer(f)\n",
    "    for v in noun_list:\n",
    "        csvw.writerow(v)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "showing info https://raw.githubusercontent.com/nltk/nltk_data/gh-pages/index.xml\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import nltk\n",
    "nltk.download()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\edu007\\.conda\\envs\\project\\lib\\site-packages\\konlpy\\tag\\_okt.py:16: UserWarning: \"Twitter\" has changed to \"Okt\" since KoNLPy v0.4.5.\n",
      "  warn('\"Twitter\" has changed to \"Okt\" since KoNLPy v0.4.5.')\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['지방공무원법', '일부', '개정', '법률', '안', '정의화', '의원', '대표', '발의', '의', '안', '번', '호', '발의', '연월일', '발', '의', '자', '정의화', '이명수', '김을동', '이사철', '여상규', '안규백', '황영철', '박영아', '김정훈', '김학송', '의원', '인', '제안', '이유', '및', '내용', '초등학교', '저학년', '경우', '부모', '사랑', '필요', '나이', '현재', '공무원', '자녀', '양육', '위', '육아휴직', '수', '자녀', '나이', '만', '세', '이하', '초등학교', '저학년', '자녀', '위', '해당', '부모님', '일자리', '곧', '출산', '의욕', '저하', '문제', '수', '것임', '따라서', '육아휴직', '자녀', '연령', '만', '세', '이하', '개정', '것임', '안', '제', '항제', '호', '법률', '제', '호', '지방공무원법', '일부', '개정', '법률', '안', '지방공무원법', '일부', '다음', '개정', '제', '항제', '호', '중', '만', '세', '이하', '초등학교', '취학', '전', '자녀', '만', '세', '이하', '취학', '중인', '경우', '초등학교', '학년', '이하', '말', '의', '자녀', '로', '부', '칙', '이', '법', '공포', '날', '시행', '신', '구조', '문대비', '표', '현', '행', '개', '정', '안', '제', '휴직', '생', '략', '제', '휴직', '현행', '공무원', '다음', '각', '호의', '느', '하나', '해당', '사유', '직', '임용', '휴직', '명', '수', '다만', '제', '호', '의', '경우', '대통령령', '정', '사정', '직', '명', '생', '략', '현행', '만', '세', '이하', '초등학교', '취', '만', '세', '이하', '취학', '중인', '경우', '학', '전', '자녀', '양육', '위', '초등학교', '학년', '이하', '여', '여자', '공무원', '말', '의', '자녀', '임신', '출산', '때', '생', '략', '생', '략', '현행', '현행', '지방공무원법', '일부', '개정', '법률', '안', '등', '비용', '추계', '첨부', '사유', '재정', '요인', '개정안', '국가공무원', '법', '제', '항제', '호', '중', '국가공무원', '육아', '휴직', '가능', '자녀', '연령', '만', '세', '이하', '만', '세', '이하', '방공', '무', '법', '제', '항제', '호', '중', '지방', '공무원', '육아휴직', '가능', '자녀', '연', '령', '만', '세', '이하', '만', '세', '이하', '교육', '공무원', '법', '제', '항', '제', '중', '교육', '공무원', '육아휴직', '가능', '자녀', '연령', '만', '세', '이하', '만', '세', '이하', '남녀', '고용', '평등', '일', '가정', '양립', '지원', '관', '법률', '제', '항', '중', '근로자', '육아휴직', '가능', '자녀', '연령', '만', '세', '이하', '만', '세', '이하', '조정', '함', '추가', '재정', '요가', '예상', '첨부', '근거', '규정', '의안', '비용', '추계', '관', '규칙', '제', '항', '단서', '중', '제', '호', '예상', '비용', '평균', '미만', '거나', '한시', '경비', '로서', '총', '미만', '경우', '해당', '함', '첨부', '사유', '개정안', '국가', '지방', '교육', '공무원', '및', '근로자', '육아휴직', '신청', '수', '자녀', '연령', '만', '세', '이하', '만', '세', '이하', '상향', '조정', '함', '추가', '재정', '요가', '예상', '법률', '개정안', '시행', '고', '가정', '경우', '현재', '자녀', '연령', '세이', '육아휴직', '신청', '국가', '지방', '교육', '공무원', '및', '근로자', '대상', '대상', '연령', '확대', '육아휴직', '신청', '수가', '정도', '것', '예상', '이', '경우', '발생', '비용', '현행법', '월', '이나', '현재', '관련', '법령', '개정', '추진', '이', '육아휴직', '지급', '월급', '액', '비례', '육아휴직', '여가', '지급', '기', '때문', '법령', '개정', '가정', '추계', '경우', '육아휴직', '급여', '액', '육아휴직', '지급', '월급', '여의', '해당', '육아휴직', '발생', '경우', '발생', '비용', '체인', '고용', '인건비', '육아휴직', '월급', '액', '이', '대비', '육아휴직', '지급', '임금', '더', '이상', '발생', '따라서', '실제', '발생', '비용', '육', '휴직', '지급', '월', '급여', '액', '연령', '확대', '발생', '비용', '인', '육아휴직', '월급', '액', '체인', '고용', '인건비', '차', '액', '이', '값', '추가', '재정', '소요', '발생', '볼', '수', '추가', '비용', '발생', '여부', '위', '비용', '대한', '리모', '델', '이', '비용', '발생', '여부', '하자', '모델', '사', '용', '변수', '다음', '정의', '발생', '비용', '육아휴직', '액', '육아휴직', '대상자', '수', '육아휴직', '대상자', '월급', '액', '육아휴직', '발생', '경우', '대체', '고용', '확률', '대체', '고용', '인력', '지급', '월급', '액', '위', '수식', '육아휴직', '액', '육아휴직', '월급', '액', '예정', '므', '육아휴직', '액', '육아휴직', '발생', '경우', '대체', '고용', '확률', '고용노동부', '육아휴직', '관련', '자료', '이용', '고용노동부', '경우', '육아휴직', '급여', '대상자', '명', '육아휴직', '체인', '고용', '상인', '명', '현황', '정리', '다음', '표', '표', '육아휴직', '급여', '수급', '수', '및', '체인', '고용', '현황', '단위', '명', '평균', '육아휴직', '급여', '수급', '체인', '채용', '비', '율', '자료', '고용노동부', '자료', '바탕', '국회예산정책처', '작성', '위', '표', '의', '자료', '육아휴직', '발생', '경우', '대체', '고용', '확률', '의', '값', '가정', '비용', '발생', '가정', '위', '의', '수식', '다시', '작성', '다음', '수식', '육아휴직', '액', '위', '수식', '육아휴직', '월', '급여', '액', '대입', '대체', '고용', '인력', '자', '지급', '월', '급여', '액', '추정', '육아휴직', '월', '원', '가정', '대체', '고용', '인력', '육아휴직', '월', '급여', '액', '배', '달', '월', '이상', '지급', '추가', '비용', '이', '발생', '대체', '고용', '인력', '육아휴직', '더', '월급', '액', '주지', '것', '배', '이상', '월급', '주지', '것', '이기', '때문', '추가', '비용', '발생', '보기', '위', '수식', '체인', '고용', '확률', '를', '로', '가정', '이', '경우', '육아휴직', '체인', '월', '이상', '지급', '추가', '비용', '발생', '행정안전부', '통계', '자료', '행정안전부', '통계', '연감', '지방', '공무원', '육', '휴직', '현황', '자료', '여기', '육아휴직', '발생', '경우', '체인', '주로', '임용', '기자', '일용직', '활용', '것', '따라서', '공무원', '경우', '표', '기업체', '체인', '고용', '확률', '것', '이상', '논의', '바탕', '육아휴직', '기간', '만', '만', '세로', '연장', '도', '법률', '개정', '추가', '비용', '발생', '것', '예상', '작성자', '국회예산정책처', '법안', '비용', '추계', '팀', '팀', '장', '정', '문', '종', '예산', '분석관', '김', '태', '완']\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "FreqDist({'\\n\\n': 127, '.': 49, '의': 46, '육아휴직': 38, '을': 28, '(': 27, ')': 26, '이': 25, '자': 24, '에': 23, ...})"
      ]
     },
     "execution_count": 97,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from konlpy.corpus import kobill   \n",
    "import nltk\n",
    "from konlpy.tag import Twitter\n",
    "files_ko = kobill.fileids()         \n",
    "doc_ko = kobill.open('1809890.txt').read()\n",
    "t = Twitter()\n",
    "tokens_ko = t.morphs(doc_ko)\n",
    "nouns = t.nouns(doc_ko)\n",
    "print(nouns)\n",
    "ko = nltk.Text(tokens_ko, name='대한민국 국회 의안 제 1809890호')   \n",
    "ko.vocab() "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "김형호 영화시장 분석가는 '1987'의 네이버 영화 정보 네티즌 10점 평에서 언급된 단어들을 지난해 12월 27일부터 올해 1월 10일까지 통계 프로그램 R과 KoNLP 패키지로 텍스트마이닝하여 분석했다.\n"
     ]
    }
   ],
   "source": [
    "from pykospacing import Spacing\n",
    "spacing = Spacing() # 띄어쓰기 교정\n",
    "print(spacing(\"김형호영화시장분석가는'1987'의네이버영화정보네티즌10점평에서언급된단어들을지난해12월27일부터올해1월10일까지통계프로그램R과KoNLP패키지로텍스트마이닝하여분석했다.\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting gensim==3.4.0Note: you may need to restart the kernel to use updated packages.\n",
      "\n",
      "  Downloading gensim-3.4.0.tar.gz (22.2 MB)\n",
      "  Preparing metadata (setup.py): started\n",
      "  Preparing metadata (setup.py): finished with status 'done'\n",
      "Requirement already satisfied: numpy>=1.11.3 in c:\\users\\edu007\\.conda\\envs\\project\\lib\\site-packages (from gensim==3.4.0) (1.19.5)\n",
      "Requirement already satisfied: scipy>=0.18.1 in c:\\users\\edu007\\.conda\\envs\\project\\lib\\site-packages (from gensim==3.4.0) (1.7.1)\n",
      "Requirement already satisfied: six>=1.5.0 in c:\\users\\edu007\\.conda\\envs\\project\\lib\\site-packages (from gensim==3.4.0) (1.15.0)\n",
      "Requirement already satisfied: smart_open>=1.2.1 in c:\\users\\edu007\\.conda\\envs\\project\\lib\\site-packages (from gensim==3.4.0) (5.2.1)\n",
      "Building wheels for collected packages: gensim\n",
      "  Building wheel for gensim (setup.py): started\n",
      "  Building wheel for gensim (setup.py): finished with status 'done'\n",
      "  Created wheel for gensim: filename=gensim-3.4.0-cp37-cp37m-win_amd64.whl size=22557541 sha256=f329b8db0e574bbc8dd6e2e9402040e9b44c56f8bbb00f5d8c5567db4c110fa5\n",
      "  Stored in directory: c:\\users\\edu007\\appdata\\local\\pip\\cache\\wheels\\de\\a4\\46\\4e18f7d25915b16e0e790a5362e455aba6cadc486994806c05\n",
      "Successfully built gensim\n",
      "Installing collected packages: gensim\n",
      "  Attempting uninstall: gensim\n",
      "    Found existing installation: gensim 3.8.3\n",
      "    Uninstalling gensim-3.8.3:\n",
      "      Successfully uninstalled gensim-3.8.3\n",
      "Successfully installed gensim-3.4.0\n"
     ]
    }
   ],
   "source": [
    "pip install gensim==3.4.0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "풀이 시인의 나는 된 위에도 많은 어머니, 아무 봅니다.그리워 나는 당신은 이웃 그리고 가난한 이름과, 멀듯이, 듯합니다.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "'\\ntext (str) – 원본 텍스트\\nratio (float, optional) – 요약할 문장 수 비율을 결정하는 0~1 사이 Float 값\\nword_count (int or None, optional) – 출력에 포함할 단어 수. ratio와 함께 사용되는 경우 ratio는 무시.\\nsplit (bool, optional) – True면 문장 list가 반환된다. False는 조인(join)된 문자열이 반환\\n'"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from gensim.summarization.summarizer import summarize # 문서 요약\n",
    "text = '이 다 무성할 같이 이네들은 패, 다하지 하나에 가득 있습니다. 내 가슴속에 별 하나에 많은 듯합니다. 내 까닭이요, 멀듯이, 비둘기, 다 노루, 거외다. 말 별을 책상을 지나고 추억과 자랑처럼 아무 불러 슬퍼하는 까닭입니다. 별이 릴케 별 별 있습니다. 보고, 벌써 벌레는 시와 무덤 듯합니다. 그리워 책상을 별들을 자랑처럼 별 소녀들의 가을로 너무나 까닭입니다. 나는 봄이 패, 불러 북간도에 무성할 나의 까닭입니다. 풀이 시인의 나는 된 위에도 많은 어머니, 아무 봅니다.그리워 나는 당신은 이웃 그리고 가난한 이름과, 멀듯이, 듯합니다. 다 된 동경과 하나에 별 같이 가슴속에 당신은 있습니다. 노루, 언덕 딴은 멀리 패, 말 까닭입니다. 말 북간도에 어머님, 강아지, 별을 계십니다. 무성할 무엇인지 시인의 잠, 벌레는 버리었습니다. 어머님, 하나의 애기 있습니다. 새워 하나의 이제 봅니다. 말 너무나 새워 이웃 아름다운 까닭이요, 많은 책상을 거외다. 까닭이요, 노루, 된 하나에 별에도 그리고 풀이 듯합니다. 너무나 벌레는 쓸쓸함과 어머님, 하나에 우는 때 다하지 이름과, 버리었습니다.이름을 묻힌 헤일 무덤 그러나 별 애기 있습니다. 이름과, 가을 별 이네들은 남은 지나가는 봄이 사랑과 계십니다. 까닭이요, 나는 이름자를 버리었습니다. 쓸쓸함과 동경과 별 까닭입니다. 이국 이네들은 하늘에는 새겨지는 있습니다. 어머니 하나에 남은 겨울이 거외다. 하나에 아침이 아스라히 봄이 어머니, 별 계십니다. 위에도 하나에 별 아무 봅니다. 시와 새워 이름을 멀듯이, 까닭이요, 까닭입니다. 나는 까닭이요, 불러 어머니 내 패, 라이너 봅니다.북간도에 내린 추억과 보고, 토끼, 때 밤이 봅니다. 내일 별 어머님, 계십니다. 새워 내일 별 계집애들의 내 가난한 계십니다. 덮어 하늘에는 별 아이들의 듯합니다. 당신은 같이 비둘기, 가득 사람들의 내일 시인의 청춘이 봅니다. 별 그리워 강아지, 부끄러운 하나에 까닭입니다. 피어나듯이 이웃 비둘기, 이런 별 슬퍼하는 못 거외다. 별 밤을 경, 이름자 프랑시스 비둘기, 않은 아무 거외다. 별 노새, 딴은 묻힌 버리었습니다. 된 위에 이름자 봅니다. 이름자 계집애들의 마리아 이런 걱정도 버리었습니다.쓸쓸함과 별 내일 묻힌 이국 이 이름과, 별 무엇인지 봅니다. 경, 내 다하지 남은 밤을 봅니다. 아직 별 가득 벌써 내일 북간도에 나는 별들을 거외다. 하나의 어머니, 헤는 둘 까닭입니다. 다 별 어머니, 별들을 아직 겨울이 아스라히 있습니다. 소녀들의 사랑과 청춘이 계십니다. 겨울이 이름자 아이들의 잠, 가을로 이름과, 지나가는 하나에 새겨지는 있습니다. 이름과, 위에 내린 헤는 무엇인지 노루, 말 마디씩 계십니다. 된 같이 계절이 있습니다. 자랑처럼 아무 당신은 가슴속에 별 오면 나의 있습니다.소녀들의 많은 까닭이요, 불러 가난한 그러나 하나에 아직 걱정도 계십니다. 이름과, 애기 라이너 계십니다. 라이너 이웃 슬퍼하는 프랑시스 묻힌 우는 까닭입니다. 다 마리아 사람들의 헤는 많은 까닭입니다. 우는 별빛이 경, 속의 이 오는 강아지, 사랑과 내일 봅니다. 벌써 어머니 속의 까닭입니다. 자랑처럼 무덤 흙으로 계십니다. 마디씩 어머니, 오면 노루, 강아지, 별이 아침이 이름을 쓸쓸함과 봅니다. 프랑시스 비둘기, 하나에 잔디가 이런 벌써 아무 지나가는 보고, 버리었습니다.강아지, 가을로 봄이 이제 봅니다. 지나고 남은 새겨지는 까닭이요, 지나가는 이름자 마리아 토끼, 듯합니다. 청춘이 하나의 동경과 별 별 별이 헤일 나는 있습니다. 이제 슬퍼하는 경, 이국 겨울이 봄이 봅니다. 흙으로 어머니 별이 패, 봄이 같이 말 풀이 계절이 까닭입니다. 별빛이 이국 별 거외다. 내 별을 아스라히 소녀들의 가난한 무덤 그리워 차 위에 버리었습니다. 무엇인지 마디씩 언덕 나의 노루, 거외다. 책상을 가을 새겨지는 한 버리었습니다. 하나의 차 가슴속에 딴은 새워 무덤 까닭입니다.하늘에는 마리아 없이 별 이름자를 우는 책상을 강아지, 까닭입니다. 묻힌 오는 언덕 우는 거외다. 내일 멀듯이, 멀리 아직 이름과, 노새, 헤일 덮어 봅니다. 이름을 아이들의 이름을 어머니, 위에 무엇인지 것은 계십니다. 나의 가을로 이름자 애기 별이 봅니다. 못 가득 슬퍼하는 노루, 사랑과 다하지 까닭입니다. 옥 된 헤일 다 있습니다. 별 오는 하나의 별 것은 별 계절이 다하지 지나고 듯합니다. 어머님, 파란 헤는 릴케 북간도에 하나에 그리워 시인의 보고, 봅니다. 둘 릴케 가을 계십니다.이름과 불러 별 흙으로 보고, 까닭입니다. 이웃 마리아 없이 자랑처럼 않은 계십니다. 이름과 무덤 그리고 까닭이요, 어머니, 가슴속에 풀이 한 듯합니다. 이런 까닭이요, 어머님, 많은 지나고 않은 멀리 있습니다. 가난한 별들을 이국 가을 있습니다. 하나에 비둘기, 오면 이런 때 언덕 봅니다. 나는 노루, 지나가는 밤을 같이 위에 듯합니다. 오면 나는 어머니, 멀듯이, 아스라히 까닭입니다. 노새, 남은 슬퍼하는 책상을 아스라히 계십니다.가을로 헤는 이국 이런 이네들은 소녀들의 새워 내 하나에 까닭입니다. 프랑시스 새겨지는 어머니, 내 쉬이 까닭입니다. 때 어머니, 둘 거외다. 쉬이 무엇인지 까닭이요, 책상을 내린 별에도 듯합니다. 하나에 나는 나는 시인의 까닭입니다. 이름을 별 별 하나에 있습니다. 묻힌 불러 계집애들의 계절이 당신은 밤이 이런 듯합니다. 내 별 당신은 새워 어머님, 이제 어머니, 듯합니다. 추억과 이 것은 위에도 까닭입니다. 차 둘 하나 없이 경, 까닭입니다.'\n",
    "#'자연어처리는 인공지능의 한 줄기 입니다. 시퀀스 투 시퀀스의 등장 이후로 딥러닝을 활용한 자연어처리는 새로운 전기를 맞이하게 되었습니다. 문장을 받아 단순히 수치로 나타내던 시절을 넘어, 원하는대로 문장을 만들어낼 수 있게 된 것 입니다. 이에따라 이전까지 큰 변화가 없었던 자연어처리 분야의 연구는 폭발적으로 늘어나기 시작하여, 곧 기계번역 시스템은 신경망에 의해 정복 당하였습니다. 또한, attention 기법의 고도화로 전이학습이 발전하면서, QA 문제도 사람보다 정확한 수준이 되었습니다.'\n",
    "print(summarize(text,word_count=10)) #   , ratio=0.1\n",
    "'''\n",
    "text (str) – 원본 텍스트\n",
    "ratio (float, optional) – 요약할 문장 수 비율을 결정하는 0~1 사이 Float 값\n",
    "word_count (int or None, optional) – 출력에 포함할 단어 수. ratio와 함께 사용되는 경우 ratio는 무시.\n",
    "split (bool, optional) – True면 문장 list가 반환된다. False는 조인(join)된 문자열이 반환\n",
    "'''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "from nltk.tokenize import sent_tokenize # 문장 분할\n",
    "text = '이 다 무성할 같이 이네들은 패, 다하지 하나에 가득 있습니다. 내 가슴속에 별 하나에 많은 듯합니다. 내 까닭이요, 멀듯이, 비둘기, 다 노루, 거외다. 말 별을 책상을 지나고 추억과 자랑처럼 아무 불러 슬퍼하는 까닭입니다. 별이 릴케 별 별 있습니다. 보고, 벌써 벌레는 시와 무덤 듯합니다. 그리워 책상을 별들을 자랑처럼 별 소녀들의 가을로 너무나 까닭입니다. 나는 봄이 패, 불러 북간도에 무성할 나의 까닭입니다. 풀이 시인의 나는 된 위에도 많은 어머니, 아무 봅니다.그리워 나는 당신은 이웃 그리고 가난한 이름과, 멀듯이, 듯합니다. 다 된 동경과 하나에 별 같이 가슴속에 당신은 있습니다. 노루, 언덕 딴은 멀리 패, 말 까닭입니다. 말 북간도에 어머님, 강아지, 별을 계십니다. 무성할 무엇인지 시인의 잠, 벌레는 버리었습니다. 어머님, 하나의 애기 있습니다. 새워 하나의 이제 봅니다. 말 너무나 새워 이웃 아름다운 까닭이요, 많은 책상을 거외다. 까닭이요, 노루, 된 하나에 별에도 그리고 풀이 듯합니다. 너무나 벌레는 쓸쓸함과 어머님, 하나에 우는 때 다하지 이름과, 버리었습니다.이름을 묻힌 헤일 무덤 그러나 별 애기 있습니다. 이름과, 가을 별 이네들은 남은 지나가는 봄이 사랑과 계십니다. 까닭이요, 나는 이름자를 버리었습니다. 쓸쓸함과 동경과 별 까닭입니다. 이국 이네들은 하늘에는 새겨지는 있습니다. 어머니 하나에 남은 겨울이 거외다. 하나에 아침이 아스라히 봄이 어머니, 별 계십니다. 위에도 하나에 별 아무 봅니다. 시와 새워 이름을 멀듯이, 까닭이요, 까닭입니다. 나는 까닭이요, 불러 어머니 내 패, 라이너 봅니다.북간도에 내린 추억과 보고, 토끼, 때 밤이 봅니다. 내일 별 어머님, 계십니다. 새워 내일 별 계집애들의 내 가난한 계십니다. 덮어 하늘에는 별 아이들의 듯합니다. 당신은 같이 비둘기, 가득 사람들의 내일 시인의 청춘이 봅니다. 별 그리워 강아지, 부끄러운 하나에 까닭입니다. 피어나듯이 이웃 비둘기, 이런 별 슬퍼하는 못 거외다. 별 밤을 경, 이름자 프랑시스 비둘기, 않은 아무 거외다. 별 노새, 딴은 묻힌 버리었습니다. 된 위에 이름자 봅니다. 이름자 계집애들의 마리아 이런 걱정도 버리었습니다.쓸쓸함과 별 내일 묻힌 이국 이 이름과, 별 무엇인지 봅니다. 경, 내 다하지 남은 밤을 봅니다. 아직 별 가득 벌써 내일 북간도에 나는 별들을 거외다. 하나의 어머니, 헤는 둘 까닭입니다. 다 별 어머니, 별들을 아직 겨울이 아스라히 있습니다. 소녀들의 사랑과 청춘이 계십니다. 겨울이 이름자 아이들의 잠, 가을로 이름과, 지나가는 하나에 새겨지는 있습니다. 이름과, 위에 내린 헤는 무엇인지 노루, 말 마디씩 계십니다. 된 같이 계절이 있습니다. 자랑처럼 아무 당신은 가슴속에 별 오면 나의 있습니다.소녀들의 많은 까닭이요, 불러 가난한 그러나 하나에 아직 걱정도 계십니다. 이름과, 애기 라이너 계십니다. 라이너 이웃 슬퍼하는 프랑시스 묻힌 우는 까닭입니다. 다 마리아 사람들의 헤는 많은 까닭입니다. 우는 별빛이 경, 속의 이 오는 강아지, 사랑과 내일 봅니다. 벌써 어머니 속의 까닭입니다. 자랑처럼 무덤 흙으로 계십니다. 마디씩 어머니, 오면 노루, 강아지, 별이 아침이 이름을 쓸쓸함과 봅니다. 프랑시스 비둘기, 하나에 잔디가 이런 벌써 아무 지나가는 보고, 버리었습니다.강아지, 가을로 봄이 이제 봅니다. 지나고 남은 새겨지는 까닭이요, 지나가는 이름자 마리아 토끼, 듯합니다. 청춘이 하나의 동경과 별 별 별이 헤일 나는 있습니다. 이제 슬퍼하는 경, 이국 겨울이 봄이 봅니다. 흙으로 어머니 별이 패, 봄이 같이 말 풀이 계절이 까닭입니다. 별빛이 이국 별 거외다. 내 별을 아스라히 소녀들의 가난한 무덤 그리워 차 위에 버리었습니다. 무엇인지 마디씩 언덕 나의 노루, 거외다. 책상을 가을 새겨지는 한 버리었습니다. 하나의 차 가슴속에 딴은 새워 무덤 까닭입니다.하늘에는 마리아 없이 별 이름자를 우는 책상을 강아지, 까닭입니다. 묻힌 오는 언덕 우는 거외다. 내일 멀듯이, 멀리 아직 이름과, 노새, 헤일 덮어 봅니다. 이름을 아이들의 이름을 어머니, 위에 무엇인지 것은 계십니다. 나의 가을로 이름자 애기 별이 봅니다. 못 가득 슬퍼하는 노루, 사랑과 다하지 까닭입니다. 옥 된 헤일 다 있습니다. 별 오는 하나의 별 것은 별 계절이 다하지 지나고 듯합니다. 어머님, 파란 헤는 릴케 북간도에 하나에 그리워 시인의 보고, 봅니다. 둘 릴케 가을 계십니다.이름과 불러 별 흙으로 보고, 까닭입니다. 이웃 마리아 없이 자랑처럼 않은 계십니다. 이름과 무덤 그리고 까닭이요, 어머니, 가슴속에 풀이 한 듯합니다. 이런 까닭이요, 어머님, 많은 지나고 않은 멀리 있습니다. 가난한 별들을 이국 가을 있습니다. 하나에 비둘기, 오면 이런 때 언덕 봅니다. 나는 노루, 지나가는 밤을 같이 위에 듯합니다. 오면 나는 어머니, 멀듯이, 아스라히 까닭입니다. 노새, 남은 슬퍼하는 책상을 아스라히 계십니다.가을로 헤는 이국 이런 이네들은 소녀들의 새워 내 하나에 까닭입니다. 프랑시스 새겨지는 어머니, 내 쉬이 까닭입니다. 때 어머니, 둘 거외다. 쉬이 무엇인지 까닭이요, 책상을 내린 별에도 듯합니다. 하나에 나는 나는 시인의 까닭입니다. 이름을 별 별 하나에 있습니다. 묻힌 불러 계집애들의 계절이 당신은 밤이 이런 듯합니다. 내 별 당신은 새워 어머님, 이제 어머니, 듯합니다. 추억과 이 것은 위에도 까닭입니다. 차 둘 하나 없이 경, 까닭입니다.'\n",
    "texts = sent_tokenize(text)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "from krwordrank.word import KRWordRank # 한국어의 특징을 반영하여 비지도학습 기반으로 한국어의 단어를 추출\n",
    "min_count = 5   # 단어의 최소 출현 빈도수 (그래프 생성 시)\n",
    "max_length = 10 # 단어의 최대 길이\n",
    "wordrank_extractor = KRWordRank(min_count=min_count, max_length=max_length)\n",
    "beta = 0.85    # PageRank의 decaying factor beta\n",
    "max_iter = 10\n",
    "keywords, rank, graph = wordrank_extractor.extract(texts, beta, max_iter)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "'''\n",
    "<< INPUT >>\n",
    "- texts : 문자열의 리스트 형태로\n",
    "<< PARAMETERS >>\n",
    "- penalty : 패널티 설정 가능\n",
    "- stopwords : 키워드에서 제거될 단어, 키워드 벡터를 만들때도 사용하지 않는다.\n",
    "- diversity : 코싸인 유사도 기준 핵심문장간의 최소 거리, 값이 클수록 다양한 문장 선택\n",
    "- num_keywords : 키워드로 추출 될 키워스 갯수를 설정.\n",
    "- num_keysents : 핵심문장으로 추출 될 문장의 갯수 설정.\n",
    "<< OUTPUT >>\n",
    "- keywords : KR-WordRank로 학습된 키워드와 랭크 값이 dict형태로\n",
    "- sents : 핵심 문장이 list of str 형식\n",
    "'''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     하나에:\t2.7633\n",
      "  까닭입니다.:\t2.4477\n",
      "      나는:\t2.2288\n",
      "   계십니다.:\t1.8297\n",
      "      이름:\t1.7393\n",
      "    봅니다.:\t1.7310\n",
      "     책상을:\t1.6371\n",
      "      많은:\t1.6113\n",
      "      어머:\t1.5366\n",
      "     노루,:\t1.5120\n",
      "   듯합니다.:\t1.4731\n",
      "      위에:\t1.4091\n",
      "    멀듯이,:\t1.3914\n",
      "      아무:\t1.3745\n",
      "   있습니다.:\t1.3638\n",
      "     보고,:\t1.3410\n",
      "      무덤:\t1.1997\n",
      "      새워:\t1.1703\n",
      "      봄이:\t1.1427\n",
      "      묻힌:\t1.1140\n",
      "      가을:\t1.1123\n",
      "      이런:\t1.1075\n",
      "    아스라히:\t1.0768\n",
      "      이웃:\t1.0725\n",
      "      불러:\t1.0394\n",
      "     당신은:\t1.0089\n",
      "      내일:\t1.0043\n",
      "      지나:\t1.0043\n",
      "     가난한:\t0.9767\n",
      "      남은:\t0.9702\n"
     ]
    }
   ],
   "source": [
    "for word, r in sorted(keywords.items(), key=lambda x:x[1], reverse=True)[:30]:\n",
    "        print('%8s:\\t%.4f' % (word, r))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "stopwords = {'어머', '아무', '이런', '지나', '남은'} # 불용어\n",
    "passwords = {word:score for word, score in sorted(\n",
    "    keywords.items(), key=lambda x:-x[1])[:300] if not (word in stopwords)}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "scan vocabs ... \n",
      "num vocabs = 247\n",
      "done\n"
     ]
    }
   ],
   "source": [
    "from krwordrank.word import summarize_with_keywords\n",
    "\n",
    "keywords = summarize_with_keywords(texts, min_count=5, max_length=10,\n",
    "    beta=0.85, max_iter=10, stopwords=stopwords, verbose=True) \n",
    "keywords = summarize_with_keywords(texts) # with default arguments"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting wordcloud\n",
      "  Downloading wordcloud-1.8.1-cp37-cp37m-win_amd64.whl (154 kB)\n",
      "Requirement already satisfied: numpy>=1.6.1 in c:\\users\\edu007\\.conda\\envs\\project\\lib\\site-packages (from wordcloud) (1.19.5)\n",
      "Requirement already satisfied: matplotlib in c:\\users\\edu007\\.conda\\envs\\project\\lib\\site-packages (from wordcloud) (3.4.3)\n",
      "Requirement already satisfied: pillow in c:\\users\\edu007\\.conda\\envs\\project\\lib\\site-packages (from wordcloud) (8.3.2)\n",
      "Requirement already satisfied: cycler>=0.10 in c:\\users\\edu007\\.conda\\envs\\project\\lib\\site-packages (from matplotlib->wordcloud) (0.10.0)\n",
      "Requirement already satisfied: kiwisolver>=1.0.1 in c:\\users\\edu007\\.conda\\envs\\project\\lib\\site-packages (from matplotlib->wordcloud) (1.3.2)\n",
      "Requirement already satisfied: pyparsing>=2.2.1 in c:\\users\\edu007\\.conda\\envs\\project\\lib\\site-packages (from matplotlib->wordcloud) (2.4.7)\n",
      "Requirement already satisfied: python-dateutil>=2.7 in c:\\users\\edu007\\.conda\\envs\\project\\lib\\site-packages (from matplotlib->wordcloud) (2.8.2)\n",
      "Requirement already satisfied: six in c:\\users\\edu007\\.conda\\envs\\project\\lib\\site-packages (from cycler>=0.10->matplotlib->wordcloud) (1.15.0)\n",
      "Installing collected packages: wordcloud\n",
      "Successfully installed wordcloud-1.8.1\n"
     ]
    }
   ],
   "source": [
    "!pip install wordcloud\n",
    "# 워드 클라우드"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "from wordcloud import WordCloud\n",
    "\n",
    "# Set your font path\n",
    "font_path = './fonts/NanumGothic.ttf'\n",
    "\n",
    "krwordrank_cloud = WordCloud(\n",
    "    font_path = font_path,\n",
    "    width = 800,\n",
    "    height = 800,\n",
    "    background_color=\"white\"\n",
    ")\n",
    "\n",
    "krwordrank_cloud = krwordrank_cloud.generate_from_frequencies(passwords)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "fig = plt.figure(figsize=(10, 10))\n",
    "plt.imshow(krwordrank_cloud, interpolation=\"bilinear\")\n",
    "plt.show()\n",
    "fig.savefig('./wordcloud.png')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['이 다 무성할 같이 이네들은 패 다하지 하나에 가득 있습니다', '내 가슴속에 별 하나에 많은 듯합니다', '내 까닭이요 멀듯이 비둘기 다 노루 거외다', '말 별을 책상을 지나고 추억과 자랑처럼 아무 불러 슬퍼하는 까닭입니다', '별이 릴케 별 별 있습니다', '보고 벌써 벌레는 시와 무덤 듯합니다', '그리워 책상을 별들을 자랑처럼 별 소녀들의 가을로 너무나 까닭입니다', '나는 봄이 패 불러 북간도에 무성할 나의 까닭입니다', '풀이 시인의 나는 된 위에도 많은 어머니 아무 봅니다 그리워 나는 당신은 이웃 그리고 가난한 이름과 멀듯이 듯합니다', '다 된 동경과 하나에 별 같이 가슴속에 당신은 있습니다', '노루 언덕 딴은 멀리 패 말 까닭입니다', '말 북간도에 어머님 강아지 별을 계십니다', '무성할 무엇인지 시인의 잠 벌레는 버리었습니다', '어머님 하나의 애기 있습니다', '새워 하나의 이제 봅니다', '말 너무나 새워 이웃 아름다운 까닭이요 많은 책상을 거외다', '까닭이요 노루 된 하나에 별에도 그리고 풀이 듯합니다', '너무나 벌레는 쓸쓸함과 어머님 하나에 우는 때 다하지 이름과 버리었습니다 이름을 묻힌 헤일 무덤 그러나 별 애기 있습니다', '이름과 가을 별 이네들은 남은 지나가는 봄이 사랑과 계십니다', '까닭이요 나는 이름자를 버리었습니다', '쓸쓸함과 동경과 별 까닭입니다', '이국 이네들은 하늘에는 새겨지는 있습니다', '어머니 하나에 남은 겨울이 거외다', '하나에 아침이 아스라히 봄이 어머니 별 계십니다', '위에도 하나에 별 아무 봅니다', '시와 새워 이름을 멀듯이 까닭이요 까닭입니다', '나는 까닭이요 불러 어머니 내 패 라이너 봅니다 북간도에 내린 추억과 보고 토끼 때 밤이 봅니다', '내일 별 어머님 계십니다', '새워 내일 별 계집애들의 내 가난한 계십니다', '덮어 하늘에는 별 아이들의 듯합니다', '당신은 같이 비둘기 가득 사람들의 내일 시인의 청춘이 봅니다', '별 그리워 강아지 부끄러운 하나에 까닭입니다', '피어나듯이 이웃 비둘기 이런 별 슬퍼하는 못 거외다', '별 밤을 경 이름자 프랑시스 비둘기 않은 아무 거외다', '별 노새 딴은 묻힌 버리었습니다', '된 위에 이름자 봅니다', '이름자 계집애들의 마리아 이런 걱정도 버리었습니다 쓸쓸함과 별 내일 묻힌 이국 이 이름과 별 무엇인지 봅니다', '경 내 다하지 남은 밤을 봅니다', '아직 별 가득 벌써 내일 북간도에 나는 별들을 거외다', '하나의 어머니 헤는 둘 까닭입니다', '다 별 어머니 별들을 아직 겨울이 아스라히 있습니다', '소녀들의 사랑과 청춘이 계십니다', '겨울이 이름자 아이들의 잠 가을로 이름과 지나가는 하나에 새겨지는 있습니다', '이름과 위에 내린 헤는 무엇인지 노루 말 마디씩 계십니다', '된 같이 계절이 있습니다', '자랑처럼 아무 당신은 가슴속에 별 오면 나의 있습니다 소녀들의 많은 까닭이요 불러 가난한 그러나 하나에 아직 걱정도 계십니다', '이름과 애기 라이너 계십니다', '라이너 이웃 슬퍼하는 프랑시스 묻힌 우는 까닭입니다', '다 마리아 사람들의 헤는 많은 까닭입니다', '우는 별빛이 경 속의 이 오는 강아지 사랑과 내일 봅니다', '벌써 어머니 속의 까닭입니다', '자랑처럼 무덤 흙으로 계십니다', '마디씩 어머니 오면 노루 강아지 별이 아침이 이름을 쓸쓸함과 봅니다', '프랑시스 비둘기 하나에 잔디가 이런 벌써 아무 지나가는 보고 버리었습니다 강아지 가을로 봄이 이제 봅니다', '지나고 남은 새겨지는 까닭이요 지나가는 이름자 마리아 토끼 듯합니다', '청춘이 하나의 동경과 별 별 별이 헤일 나는 있습니다', '이제 슬퍼하는 경 이국 겨울이 봄이 봅니다', '흙으로 어머니 별이 패 봄이 같이 말 풀이 계절이 까닭입니다', '별빛이 이국 별 거외다', '내 별을 아스라히 소녀들의 가난한 무덤 그리워 차 위에 버리었습니다', '무엇인지 마디씩 언덕 나의 노루 거외다', '책상을 가을 새겨지는 한 버리었습니다', '하나의 차 가슴속에 딴은 새워 무덤 까닭입니다 하늘에는 마리아 없이 별 이름자를 우는 책상을 강아지 까닭입니다', '묻힌 오는 언덕 우는 거외다', '내일 멀듯이 멀리 아직 이름과 노새 헤일 덮어 봅니다', '이름을 아이들의 이름을 어머니 위에 무엇인지 것은 계십니다', '나의 가을로 이름자 애기 별이 봅니다', '못 가득 슬퍼하는 노루 사랑과 다하지 까닭입니다', '옥 된 헤일 다 있습니다', '별 오는 하나의 별 것은 별 계절이 다하지 지나고 듯합니다', '어머님 파란 헤는 릴케 북간도에 하나에 그리워 시인의 보고 봅니다', '둘 릴케 가을 계십니다 이름과 불러 별 흙으로 보고 까닭입니다', '이웃 마리아 없이 자랑처럼 않은 계십니다', '이름과 무덤 그리고 까닭이요 어머니 가슴속에 풀이 한 듯합니다', '이런 까닭이요 어머님 많은 지나고 않은 멀리 있습니다', '가난한 별들을 이국 가을 있습니다', '하나에 비둘기 오면 이런 때 언덕 봅니다', '나는 노루 지나가는 밤을 같이 위에 듯합니다', '오면 나는 어머니 멀듯이 아스라히 까닭입니다', '노새 남은 슬퍼하는 책상을 아스라히 계십니다 가을로 헤는 이국 이런 이네들은 소녀들의 새워 내 하나에 까닭입니다', '프랑시스 새겨지는 어머니 내 쉬이 까닭입니다', '때 어머니 둘 거외다', '쉬이 무엇인지 까닭이요 책상을 내린 별에도 듯합니다', '하나에 나는 나는 시인의 까닭입니다', '이름을 별 별 하나에 있습니다', '묻힌 불러 계집애들의 계절이 당신은 밤이 이런 듯합니다', '내 별 당신은 새워 어머님 이제 어머니 듯합니다', '추억과 이 것은 위에도 까닭입니다', '차 둘 하나 없이 경 까닭입니다']\n"
     ]
    }
   ],
   "source": [
    "from krwordrank.hangle import normalize\n",
    "\n",
    "texts = [normalize(text, english=True, number=True) for text in texts] # 한글, 영어, 숫자를 제외한 다른 글자를 제거 \n",
    "# input data : list of str # normalize 특수기호 제거 전처리 \n",
    "print(texts)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Key-sentence extraction \n",
    "'''\n",
    " KR-WordRank 에서는 keywords 를 많이 포함한 문장을 핵심 문장으로 선택합니다. \n",
    " 문장을 추출하는 원리는 추출된 키워드의 랭크값을 이용하여 키워드 벡터를 만든 뒤, \n",
    " 코싸인 유사도 기준으로 입력된 문장 벡터가 키워드 벡터와 유사한 문장을 선택하는 것\n",
    " '''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "from krwordrank.sentence import summarize_with_sentences\n",
    "# texts 리스트 형식\n",
    "keywords, sents = summarize_with_sentences(texts, num_keywords=100, num_keysents=10) # KR-WordRank 를 학습하여 키워드와 이를 이용한 핵심 문장을 선택"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'하나에': 2.476438561944873, '까닭': 2.383691765572484, '계십니다': 2.118303720532178, '이름': 1.95492851198616, '나는': 1.9101929427274764, '봅니다': 1.836248911244121, '어머': 1.7275324825823466, '있습니다': 1.6039502138817716, '책상을': 1.4289281917762002, '많은': 1.3683528274712784, '노루': 1.2858197128548565, '듯합니다': 1.2255086682680911, '위에': 1.2095611370503634, '아무': 1.1879364380886321, '멀듯이': 1.1337540789127127, '가을': 1.129792684777726, '불러': 1.1273066366944557, '무덤': 1.1206670597426411, '보고': 1.1193526870301629, '강아지': 1.0892095538580326, '이웃': 1.046551501058747, '새워': 1.0361634932988433, '아스라히': 1.0321666208568159, '내일': 1.0027086142383324, '지나': 1.0001776901025048, '묻힌': 0.9760412197615316, '이런': 0.9632240259013721, '봄이': 0.9546645450131044, '당신은': 0.9442469487434841, '남은': 0.9320489248451124, '소녀들의': 0.9309238708234845, '버리었습니다': 0.906730645132226, '거외다': 0.8761450213960617, '그리워': 0.8353925870491595, '가난한': 0.832026194054448, '새겨지는': 0.7644541011187966, '시인의': 0.7598814860212346, '무엇인지': 0.7472586089793299, '같이': 0.7221732051359157, '우는': 0.7143408831742267, '마리': 0.6884808203654035, '다하지': 0.6794781736916912, '슬퍼': 0.6723622699248794, '헤는': 0.6626981641993915, '별이': 0.6580270177990154, '비둘': 0.6523781920638407, '가슴': 0.6257891758176384, '북간': 0.6231564359737203, '자랑처럼': 0.6214499973046228, '이국': 0.5585314964543746}\n"
     ]
    }
   ],
   "source": [
    "print(keywords)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['풀이 시인의 나는 된 위에도 많은 어머니 아무 봅니다 그리워 나는 당신은 이웃 그리고 가난한 이름과 멀듯이 듯합니다', '노새 남은 슬퍼하는 책상을 아스라히 계십니다 가을로 헤는 이국 이런 이네들은 소녀들의 새워 내 하나에 까닭입니다', '자랑처럼 아무 당신은 가슴속에 별 오면 나의 있습니다 소녀들의 많은 까닭이요 불러 가난한 그러나 하나에 아직 걱정도 계십니다', '프랑시스 비둘기 하나에 잔디가 이런 벌써 아무 지나가는 보고 버리었습니다 강아지 가을로 봄이 이제 봅니다', '너무나 벌레는 쓸쓸함과 어머님 하나에 우는 때 다하지 이름과 버리었습니다 이름을 묻힌 헤일 무덤 그러나 별 애기 있습니다', '하나의 차 가슴속에 딴은 새워 무덤 까닭입니다 하늘에는 마리아 없이 별 이름자를 우는 책상을 강아지 까닭입니다', '나는 까닭이요 불러 어머니 내 패 라이너 봅니다 북간도에 내린 추억과 보고 토끼 때 밤이 봅니다', '둘 릴케 가을 계십니다 이름과 불러 별 흙으로 보고 까닭입니다', '어머님 파란 헤는 릴케 북간도에 하나에 그리워 시인의 보고 봅니다', '이름자 계집애들의 마리아 이런 걱정도 버리었습니다 쓸쓸함과 별 내일 묻힌 이국 이 이름과 별 무엇인지 봅니다']\n"
     ]
    }
   ],
   "source": [
    "print(sents) # 핵심 문장이 list of str 형식"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 몇 가지 패러매터를 추가할 수 있습니다. 길이가 지나치게 길거나 짧은 문장을 제거하기 위해 penalty 함수를 정의합니다. \n",
    "# 아래는 길이가 25 글자부터 80 글자인 문장을 선호한다는 의미\n",
    "penalty = lambda x:0 if (25 <= len(x) <= 80) else 1\n",
    "stopwords = {'거외다', '어머', '아무', '마리'}\n",
    "\n",
    "keywords, sents = summarize_with_sentences(\n",
    "    texts,\n",
    "    penalty=penalty,\n",
    "    stopwords = stopwords,\n",
    "    diversity=0.5,\n",
    "    num_keywords=100,\n",
    "    num_keysents=10,\n",
    "    verbose=False\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'하나에': 2.476438561944873, '까닭': 2.383691765572484, '계십니다': 2.118303720532178, '이름': 1.95492851198616, '나는': 1.9101929427274764, '봅니다': 1.836248911244121, '있습니다': 1.6039502138817716, '책상을': 1.4289281917762002, '많은': 1.3683528274712784, '노루': 1.2858197128548565, '듯합니다': 1.2255086682680911, '위에': 1.2095611370503634, '멀듯이': 1.1337540789127127, '가을': 1.129792684777726, '불러': 1.1273066366944557, '무덤': 1.1206670597426411, '보고': 1.1193526870301629, '강아지': 1.0892095538580326, '이웃': 1.046551501058747, '새워': 1.0361634932988433, '아스라히': 1.0321666208568159, '내일': 1.0027086142383324, '지나': 1.0001776901025048, '묻힌': 0.9760412197615316, '이런': 0.9632240259013721, '봄이': 0.9546645450131044, '당신은': 0.9442469487434841, '남은': 0.9320489248451124, '소녀들의': 0.9309238708234845, '버리었습니다': 0.906730645132226, '그리워': 0.8353925870491595, '가난한': 0.832026194054448, '새겨지는': 0.7644541011187966, '시인의': 0.7598814860212346, '무엇인지': 0.7472586089793299, '같이': 0.7221732051359157, '우는': 0.7143408831742267, '다하지': 0.6794781736916912, '슬퍼': 0.6723622699248794, '헤는': 0.6626981641993915, '별이': 0.6580270177990154, '비둘': 0.6523781920638407, '가슴': 0.6257891758176384, '북간': 0.6231564359737203, '자랑처럼': 0.6214499973046228, '이국': 0.5585314964543746}\n"
     ]
    }
   ],
   "source": [
    "print(keywords)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "import gensim # 단어 유사도\n",
    "model = gensim.models.Word2Vec.load('./keyword_ko/ko.bin')\n",
    "'''\n",
    "# Word2Vec 유사단어\n",
    "Word2Vec에는 CBOW(Continuous Bag of Words)와 Skip-Gram 두 가지 방식\n",
    "CBOW는 주변에 있는 단어들을 가지고, 중간에 있는 단어들을 예측하는 방법입니다. \n",
    "반대로, Skip-Gram은 중간에 있는 단어로 주변 단어들을 예측하는 방법입니다.\n",
    "'''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[('고양이', 0.7290452718734741), ('거위', 0.7185635566711426), ('토끼', 0.7056223154067993), ('멧돼지', 0.6950401067733765), ('엄마', 0.6934334635734558), ('난쟁이', 0.6806551218032837), ('한마리', 0.6770296096801758), ('아가씨', 0.6750352382659912), ('아빠', 0.6729634404182434), ('목걸이', 0.6512460708618164)]\n"
     ]
    }
   ],
   "source": [
    "result = model.wv.most_similar(\"강아지\")\n",
    "print(result)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0        2021년 대기업 상생협력 지원사업(현대백화점-더 성장 마켓) 참여기업 모집 공고...\n",
      "1        2021년 산업기술R&D 연구기획사업(에너지분야) 신규지원 대상과제 공고 지원사업...\n",
      "2        2021년 2차 에너지기술개발사업 신규지원 대상 연구개발과제 공고 지원사업 상세 ...\n",
      "3        2021년 베트남 호치민 하드웨어 수동공구 전시회 한국관 참가기업 모집 공고 지원...\n",
      "4        2021년 중국 상하이 마린텍 차이나 전시회 한국관 참가기업 추가 모집 공고 지원...\n",
      "                              ...                        \n",
      "1435     [강원] 강릉시 2021년 기업경영정책자금 융자추천 계획 공고 지원사업 상세 분류...\n",
      "1436     [인천] 남동구 2021년 중소기업 통ㆍ번역 지원사업 참가업체 모집 공고 지원사업...\n",
      "1437     [제주] 경영안정 지원자금(2021년 중소기업육성자금 융자지원계획 공고) 지원사업...\n",
      "1438     [제주] 2021년 제주공동물류센터 지원사업 공고 지원사업 상세 분류체계, 신청기...\n",
      "1439     [충남] 혁신형자금(2021년 중소기업육성자금 융자 지원계획 공고) 지원사업 상세...\n",
      "Name: content, Length: 1440, dtype: object\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "df = pd.read_csv('./test.csv')\n",
    "print(df.content)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 문서 분류 https://wikidocs.net/102705\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "from sklearn.preprocessing import normalize\n",
    "from sklearn.cluster import KMeans\n",
    "\n",
    "n_clusters = 20  #20개 군집화\n",
    "\n",
    "# CountVectrizer 토큰화 (df.text)\n",
    "vectorizer = CountVectorizer()\n",
    "X = vectorizer.fit_transform(df.content) # WOB 생성\n",
    "\n",
    "# l2 정규화 적용\n",
    "X = normalize(X)  #WOB 정규화\n",
    "\n",
    "# k-means 알고리즘 \n",
    "kmeans = KMeans(n_clusters=n_clusters).fit(X) #WOB K-Means 적용\n",
    "\n",
    "# 학습된 레이블링 \n",
    "labels = kmeans.labels_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[13 14 14 ...  5  0 11] 1440\n"
     ]
    }
   ],
   "source": [
    "print(labels,len(labels))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['labels'] = labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 유사단어 Text Similarity\n",
    "#자카드 유사도 / 유클리디언 유사도 / 맨하탄 유사도 / 코사인 유사도"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "sentence = ( \"휴일 인 오늘 도 서쪽 을 중심 으로 폭염 이 이어졌는데요, 내일 은 반가운 비 소식 이 있습니다.\",\"폭염 을 피해서 휴일 에 놀러왔다가 갑작스런 비 로 인해 망연자실 하고 있습니다.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\edu007\\.conda\\envs\\project\\lib\\site-packages\\sklearn\\utils\\deprecation.py:87: FutureWarning: Function get_feature_names is deprecated; get_feature_names is deprecated in 1.0 and will be removed in 1.2. Please use get_feature_names_out instead.\n",
      "  warnings.warn(msg, category=FutureWarning)\n"
     ]
    }
   ],
   "source": [
    "# tf-idf\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "\n",
    "# 객체 생성\n",
    "tfidf_vectorizer = TfidfVectorizer()\n",
    "# 문장 벡터화 진행\n",
    "tfidf_matrix = tfidf_vectorizer.fit_transform(sentence)\n",
    "# 각 단어\n",
    "text = tfidf_vectorizer.get_feature_names()\n",
    "# 각 단어의 벡터 값\n",
    "idf = tfidf_vectorizer.idf_\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "문장 1의 집합 =  {'을', '이', '내일', '반가운', '휴일', '비', '소식', '인', '중심', '폭염', '있습니다.', '도', '으로', '오늘', '이어졌는데요,', '은', '서쪽'}\n",
      "문장 2의 집합 =  {'을', '피해서', '휴일', '비', '폭염', '있습니다.', '놀러왔다가', '하고', '인해', '갑작스런', '에', '로', '망연자실'}\n",
      "합집합 =  {'이', '에', '비', '인', '폭염', '도', '갑작스런', '으로', '로', '은', '을', '내일', '반가운', '피해서', '휴일', '소식', '중심', '있습니다.', '놀러왔다가', '하고', '인해', '오늘', '이어졌는데요,', '망연자실', '서쪽'}\n",
      "교집합 =  {'을', '휴일', '비', '있습니다.', '폭염'}\n",
      "자카드 유사도 =  0.2\n"
     ]
    }
   ],
   "source": [
    "# 자카드 유사도( Jaccard Similarity)\n",
    "# 두 문장을 각각 단어의 집합으로 만든 뒤 두 집합을 통해 유사도를 측정하는 방식 중 하나\n",
    "# 유사도를 측정하는 방법은 두 집합의 교집합인 공통된 단어의 개수를 두 집합의 합집합, 전체 단어의 갯수로 나눈다.\n",
    "# 결괏값은 공통의 원소의 개수에 따라 0과 1사이의 값이 나올 것이고, 1에 가까울수록 유사도가 높다\n",
    "\n",
    "sent = (\"휴일 인 오늘 도 서쪽 을 중심 으로 폭염 이 이어졌는데요, 내일 은 반가운 비 소식 이 있습니다.\", \n",
    "        \"폭염 을 피해서 휴일 에 놀러왔다가 갑작스런 비 로 인해 망연자실 하고 있습니다.\") \n",
    "        \n",
    "tokenized_doc1 = set(sent[0].split(' '))\n",
    "tokenized_doc2 = set(sent[1].split(' '))\n",
    "print(\"문장 1의 집합 = \", tokenized_doc1)\n",
    "print(\"문장 2의 집합 = \", tokenized_doc2)\n",
    "\n",
    "union = set(tokenized_doc1).union(set(tokenized_doc2))\n",
    "print(\"합집합 = \", union)\n",
    "\n",
    "intersection = set(tokenized_doc1).intersection(set(tokenized_doc2))\n",
    "print(\"교집합 = \", intersection)\n",
    "\n",
    "jaccardScore = len(intersection)/len(union)\n",
    "\n",
    "print(\"자카드 유사도 = \", jaccardScore)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.17952266]])"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 코사인 유사도\n",
    "# 코사인 유사도는 두 개의 벡터값에서 코사인 각도를 구하는 방법이다. \n",
    "# 코사인 유사도 값은 -1과 1사이의 값을 가지고 1에 가까울수록 유사하다는 것을 의미한다.\n",
    "from sklearn.metrics.pairwise import cosine_similarity\n",
    "cosine_similarity(tfidf_matrix[0:1], tfidf_matrix[1:2])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.20491229]])"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 유클리디언 유사도 ( L2 Distance )\n",
    "# 가장 기본적인 거리를 측정하는 유사도 공식이다.\n",
    "# 유클리디언 거리 = L2 거리 : N차원 공간에서 두 점 사이의 최단 거리를 구하는 접근법\n",
    "# 유클리디언 유사도는 단순히 두 점 사이의 거리를 뜻하므로 값에 제한이 없다.\n",
    "# 다른 유사도와 비교하려면 0과 1사이의 값으로 맞춰주어야한다.\n",
    "from sklearn.metrics.pairwise import euclidean_distances\n",
    "import numpy as np\n",
    "\n",
    "def l1_normalize(v): # 정규화\n",
    "    norm = np.sum(v)\n",
    "    return v / norm\n",
    "    \n",
    "    \n",
    "tfidf_norm_l1 = l1_normalize(tfidf_matrix)\n",
    "euclidean_distances(tfidf_norm_l1[0:1], tfidf_norm_l1[1:2])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 맨하탄 유사도 ( L1 Distance )\n",
    "# 맨하탄 거리를 통해 유사도를 측정하는 방법이다.\n",
    "# 사각형 격자로 이뤄진 지도에서 출발점에서 도착점까지를 가로지르지 않고 갈 수 있는 최단거리를 구하는 공식\n",
    "# 맨하탄 유사도  또한 유클리디언 유사도와 마찬가지로 값이 계속 커질 수 있다.\n",
    "# 따라서 0과 1 사이의 값을 갖도록 L1 정규화방법을 사용한 벡터 값으로 유사도를 측정한다.\n",
    "import numpy as np\n",
    "\n",
    "def l1_normalize(v):\n",
    "    norm = np.sum(v)\n",
    "    return v / norm \n",
    "  \n",
    "# L1 정규화  \n",
    "tfidf_norm_l1 = l1_normalize(tfidf_matrix)\n",
    "\n",
    "# 맨하탄 유사도\n",
    "from sklearn.metrics.pairwise import manhattan_distances\n",
    "\n",
    "manhattan_distances(tfidf_norm_l1[0:1], tfidf_norm_l1[1:2])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-4.48758937e-02, -5.26038110e-01,  2.18058324e+00, -2.70098805e-01,\n",
       "       -1.22612631e+00, -4.35174644e-01,  8.78038049e-01,  1.78903091e+00,\n",
       "        5.13344407e-01,  8.00866067e-01,  1.33781850e-01,  4.23676372e-01,\n",
       "        6.25718057e-01, -1.49838462e-01, -2.65716583e-01, -1.18363452e+00,\n",
       "        1.09828494e-01,  6.56335890e-01,  9.90437388e-01, -2.88908720e-01,\n",
       "       -4.82253104e-01, -1.71825096e-01, -5.22048175e-01, -1.37081638e-01,\n",
       "        4.01715070e-01, -3.54384005e-01, -4.38563734e-01,  3.37939188e-02,\n",
       "       -4.21380460e-01, -2.11324722e-01,  5.34786046e-01,  5.26668072e-01,\n",
       "       -3.78847629e-01, -1.59311071e-01,  1.77993524e+00, -1.28354877e-01,\n",
       "        4.53990996e-01,  6.38647795e-01,  3.22042465e-01, -6.49676397e-02,\n",
       "       -9.30086493e-01,  6.84082747e-01, -1.41875729e-01, -1.49376774e+00,\n",
       "        1.44657120e-03,  4.33810472e-01, -7.31874406e-02,  7.40463585e-02,\n",
       "       -6.44519210e-01,  1.11461449e+00, -8.20367783e-02, -5.76601565e-01,\n",
       "        1.71026981e+00,  5.43356836e-01, -1.38096079e-01, -7.13480115e-01,\n",
       "        1.37917042e-01,  1.31351754e-01, -1.64170414e-01,  5.08485079e-01,\n",
       "        1.03430068e+00, -5.08647382e-01,  3.78100970e-03,  4.74166781e-01,\n",
       "        5.89607954e-01,  3.45575511e-01,  8.35864961e-01,  8.41770291e-01,\n",
       "        9.89145637e-01, -2.74089158e-01,  1.25443026e-01,  1.02799308e+00,\n",
       "       -4.82123911e-01, -1.69151232e-01, -2.85922945e-01, -8.49736392e-01,\n",
       "        8.46186206e-02, -1.81830883e+00, -5.13923943e-01, -2.16713548e-01,\n",
       "       -1.31658185e+00, -1.02243495e+00,  8.11321557e-01,  2.19687596e-02,\n",
       "       -6.71579957e-01, -5.98710179e-01, -7.66289771e-01, -1.32737505e+00,\n",
       "       -6.29475042e-02, -1.27141750e+00, -9.70623434e-01, -2.50643134e-01,\n",
       "        5.53779125e-01, -9.36333358e-01, -3.81229192e-01,  9.78302062e-02,\n",
       "       -1.73168528e+00,  2.34338660e-02,  5.90916038e-01,  1.10269034e+00,\n",
       "       -1.77305982e-01,  5.86390793e-01, -2.25778148e-01,  1.85514867e-01,\n",
       "       -8.05007041e-01,  5.64193904e-01, -9.69711125e-01, -3.67523193e-01,\n",
       "       -1.23882771e+00, -4.01369512e-01, -1.19125068e-01, -4.65394631e-02,\n",
       "       -2.21846491e-01,  1.82357505e-02,  4.56880897e-01, -2.14156687e-01,\n",
       "       -3.06043118e-01, -1.71571866e-01,  6.13820255e-01, -4.26191688e-01,\n",
       "        5.79584062e-01,  1.03633344e-01,  6.82440162e-01, -1.95503905e-01,\n",
       "        5.96677661e-01,  6.77304804e-01, -1.25491381e+00,  3.14139165e-02,\n",
       "        1.33252859e+00, -1.52929530e-01, -5.01865447e-01,  4.62395728e-01,\n",
       "        6.49050236e-01,  4.93880659e-01,  1.26435041e-01,  3.00690591e-01,\n",
       "        3.55630815e-01, -2.70101190e-01,  5.62555134e-01, -1.03386605e+00,\n",
       "       -3.14555407e-01,  3.89042318e-01,  1.03528607e+00, -4.63004947e-01,\n",
       "       -6.14613831e-01, -2.57924795e-01, -1.05875695e+00,  1.89552337e-01,\n",
       "        5.63951731e-01,  1.79895550e-01,  2.87300587e-01, -1.12558770e+00,\n",
       "       -1.56612837e+00,  6.86075389e-01,  6.99334323e-01, -3.55547547e-01,\n",
       "        6.66430950e-01, -2.01317817e-01,  9.82120037e-01,  1.48326457e-02,\n",
       "        5.74257970e-01,  5.32084644e-01, -5.38365960e-01,  1.24933779e-01,\n",
       "       -3.11034471e-01,  5.36548980e-02,  4.54503685e-01,  1.85251966e-01,\n",
       "        1.31985113e-01,  2.39911228e-01, -9.13004637e-01, -1.16619909e+00,\n",
       "        8.83497417e-01,  3.04435015e-01, -7.70304739e-01, -7.39239752e-02,\n",
       "       -1.10028563e-02,  5.21843016e-01, -9.24460530e-01,  5.88730514e-01,\n",
       "        6.41907156e-02, -5.43511808e-01, -2.87139893e-01,  2.77004838e-01,\n",
       "       -4.67248596e-02, -5.69744051e-01, -6.73322231e-02,  3.29014838e-01,\n",
       "       -6.28687918e-01, -5.23731291e-01,  6.91283345e-01, -1.03363819e-01,\n",
       "       -7.24156022e-01, -5.24323702e-01, -5.47199488e-01,  2.67527951e-03,\n",
       "        5.74104823e-02, -1.35755897e+00,  4.52745229e-01, -3.91733170e-01],\n",
       "      dtype=float32)"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from gensim.models import Word2Vec\n",
    "model = Word2Vec.load(\"./keyword_ko/ko.bin\")\n",
    "model.wv['안녕']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3.4.0\n"
     ]
    }
   ],
   "source": [
    "import gensim\n",
    "print(gensim.__version__)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\edu007\\.conda\\envs\\project\\lib\\site-packages\\gensim\\utils.py:1197: UserWarning: detected Windows; aliasing chunkize to chunkize_serial\n",
      "  warnings.warn(\"detected Windows; aliasing chunkize to chunkize_serial\")\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[(\"('파리','Noun')\", 0.6150658130645752),\n",
       " (\"('브뤼셀','Noun')\", 0.46250566840171814),\n",
       " (\"('르펜','Noun')\", 0.46202123165130615),\n",
       " (\"('국민전선','Noun')\", 0.4543028175830841),\n",
       " (\"('베를린','Noun')\", 0.43678873777389526),\n",
       " (\"('에마뉘엘','Noun')\", 0.4343237280845642),\n",
       " (\"('멜랑숑','Noun')\", 0.4231666028499603),\n",
       " (\"('샹젤리제','Noun')\", 0.42176365852355957),\n",
       " (\"('니스','Noun')\", 0.42036759853363037),\n",
       " (\"('마린','Noun')\", 0.4149590730667114)]"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from gensim.models.keyedvectors import KeyedVectors\n",
    "pos_vectors = KeyedVectors.load_word2vec_format('pos.vec', binary=False)\n",
    "pos_vectors.most_similar(\"('대통령','Noun')\")\n",
    "pos_vectors.most_similar(positive=[\"('도쿄','Noun')\", \"('프랑스','Noun')\"], negative=[\"('일본','Noun')\"])\n"
   ]
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "29feeedf21dd36998d4d40f99975547bb5b2b526a0690d77b1e0641041067138"
  },
  "kernelspec": {
   "display_name": "Python 3.7.11 64-bit ('project': conda)",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}