03_02_nlp_and_EDA_.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "03_02 nlp and EDA .ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/jaygo-kr/brunch/blob/master/03_02_nlp_and_EDA_.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "riIfUIjOb46e",
        "colab_type": "text"
      },
      "source": [
        "# 드라마 대본 자연어 태깅 및 EDA\n",
        "\n",
        "> 들여쓴 블록\n",
        "\n",
        "\n",
        "* 이전 노트에서 정제한 대본 자료의 문장들을 테깅 해보고 데이터 분석을 통해 데터를 둘러보기\n",
        "1. 정제한 대본 파일 불러오기\n",
        "2. 한국어 태깅 시키기\n",
        "3. 명사 워드클라우드 그리기\n",
        "4. 자연어 통계분석"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6uDvTrhYf6Mn",
        "colab_type": "code",
        "outputId": "a79e25ef-00cc-427f-e486-fa0b7fdfbe74",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 55
        }
      },
      "source": [
        "import os\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "from google.colab import files, drive\n",
        "drive.mount('/content/gdrive')"
      ],
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Drive already mounted at /content/gdrive; to attempt to forcibly remount, call drive.mount(\"/content/gdrive\", force_remount=True).\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wnP-VfWJf7gd",
        "colab_type": "code",
        "outputId": "25b67ee1-ed0b-498d-87e3-493973dec3f6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 341
        }
      },
      "source": [
        "os.listdir('./gdrive/My Drive/share/suwon study/script/JAY완료/')"
      ],
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['W+제11회+대본.csv',\n",
              " 'W+제12회+대본.csv',\n",
              " 'W+제13회+대본.csv',\n",
              " 'W+제15회+대본.csv',\n",
              " 'W+제4회+대본.csv',\n",
              " 'W+제1회+대본.csv',\n",
              " 'W+제14회+대본.csv',\n",
              " 'W+제3회+대본.csv',\n",
              " 'W+제16회+대본.csv',\n",
              " 'W+제2회+대본.csv',\n",
              " 'W+제5회+대본.csv',\n",
              " 'W+제7회+대본.csv',\n",
              " 'W+제6회+대본.csv',\n",
              " 'W+제9회+대본.csv',\n",
              " 'W+제8회+대본.csv',\n",
              " 'W+제10회+대본.csv',\n",
              " 'W+제1회+대본.gsheet',\n",
              " 'text to csv.ipynb의 사본']"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 39
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LpcWVtOLgNjm",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "DATA_PATH = './gdrive/My Drive/share/suwon study/script/JAY완료/'"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9yVtgxiZkL8U",
        "colab_type": "text"
      },
      "source": [
        "## 데이터 불러오기"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5Pk3TDTggHCg",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "script_df = pd.read_csv(DATA_PATH + 'W+제1회+대본.csv')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3R8f20LcgU69",
        "colab_type": "code",
        "outputId": "6340f2da-9c75-46be-c40a-c76f925b901d",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        }
      },
      "source": [
        "script_df.head()"
      ],
      "execution_count": 42,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>act</th>\n",
              "      <th>content</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>씬/1</td>\n",
              "      <td>아테네 올림픽 자료화면 인서트 \\n</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>지문</td>\n",
              "      <td>자막 – 2004년 8월. 아테네 올림픽\\n한국 국가대표 경기 생중계 자료화면.\\n...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>씬/2</td>\n",
              "      <td>방송 스튜디오 (과거, 밤)\\n</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>지문</td>\n",
              "      <td>올림픽 중계 스튜디오. 아테네 올림픽 엠블럼과 마스코트를 배경으로 남자 MC가 진행...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>MC</td>\n",
              "      <td>네, 잠시 종목을 바꿔보겠습니다. 지금 사격 경기장에서는 50미터 권총 남자 결선이...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   act                                            content\n",
              "0  씬/1                                아테네 올림픽 자료화면 인서트 \\n\n",
              "1   지문  자막 – 2004년 8월. 아테네 올림픽\\n한국 국가대표 경기 생중계 자료화면.\\n...\n",
              "2  씬/2                                  방송 스튜디오 (과거, 밤)\\n\n",
              "3   지문  올림픽 중계 스튜디오. 아테네 올림픽 엠블럼과 마스코트를 배경으로 남자 MC가 진행...\n",
              "4   MC  네, 잠시 종목을 바꿔보겠습니다. 지금 사격 경기장에서는 50미터 권총 남자 결선이..."
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 42
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mLlssRrjkPVw",
        "colab_type": "text"
      },
      "source": [
        "## 문장 태깅하기\n",
        "* content 전체에 대하여 명사로 태깅해보자"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nkH69VakgV4E",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# 한글 태깅을 위한 패키지 설치\n",
        "! pip install konlpy"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Z_j-8i3NiJp4",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from konlpy.tag import Komoran, Twitter, Hannanum"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LAw8NNj6jLtz",
        "colab_type": "code",
        "outputId": "017cb513-b7d0-4c9f-dd36-9d93d7c3571e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 73
        }
      },
      "source": [
        "Komoran = Komoran()\n",
        "Twitter = Twitter()\n",
        "Hannanum = Hannanum()"
      ],
      "execution_count": 45,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-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": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vTjaWxFwjOK0",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# 태그 전 대본 대사 마지막에 있는 개행문자 제거\n",
        "content = script_df.content.map(lambda x: x.rstrip('\\n'))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vVLhriuto3lb",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# 태깅 입력을 위해 하나의 문장으로 연결\n",
        "content_tie = content.str.cat(sep=' ')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gbfzqHY6pXcy",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# 명사 태깅을 위한 클래스들 비교를 위해 3가지 클래스 사용\n",
        "content_nouns_ha = Hannanum.nouns(content_tie)\n",
        "content_nouns_tw = Twitter.nouns(content_tie)\n",
        "content_nouns_ko = Komoran.nouns(content_tie)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TVMetwsVqIeh",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "content_nouns_ha_pd = pd.Series(content_nouns_ha)\n",
        "content_nouns_tw_pd = pd.Series(content_nouns_tw)\n",
        "content_nouns_ko_pd = pd.Series(content_nouns_ko)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zsnWV3DlrPIx",
        "colab_type": "code",
        "outputId": "0508652b-a707-476c-dd93-617216bb327f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 359
        }
      },
      "source": [
        "# 태깅 클래스마다 차이가 있다.\n",
        "content_nouns_compare = pd.DataFrame()\n",
        "content_nouns_compare = pd.concat([content_nouns_compare, content_nouns_ko_pd.value_counts()[:30].reset_index()], axis=1)\n",
        "content_nouns_compare = pd.concat([content_nouns_compare, content_nouns_tw_pd.value_counts()[:30].reset_index()], axis=1)\n",
        "content_nouns_compare = pd.concat([content_nouns_compare, content_nouns_ha_pd.value_counts()[:30].reset_index()], axis=1)\n",
        "content_nouns_compare.columns = ['Komoran','count','Twitter','count','Hannanum','count']\n",
        "content_nouns_compare.head(10)"
      ],
      "execution_count": 50,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>Komoran</th>\n",
              "      <th>count</th>\n",
              "      <th>Twitter</th>\n",
              "      <th>count</th>\n",
              "      <th>Hannanum</th>\n",
              "      <th>count</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>강철</td>\n",
              "      <td>137</td>\n",
              "      <td>강철</td>\n",
              "      <td>137</td>\n",
              "      <td>강철</td>\n",
              "      <td>116</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>연주</td>\n",
              "      <td>109</td>\n",
              "      <td>연주</td>\n",
              "      <td>103</td>\n",
              "      <td>연주</td>\n",
              "      <td>101</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>!!</td>\n",
              "      <td>48</td>\n",
              "      <td>안</td>\n",
              "      <td>40</td>\n",
              "      <td>것</td>\n",
              "      <td>65</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>말</td>\n",
              "      <td>40</td>\n",
              "      <td>거</td>\n",
              "      <td>40</td>\n",
              "      <td>수</td>\n",
              "      <td>44</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>거</td>\n",
              "      <td>38</td>\n",
              "      <td>말</td>\n",
              "      <td>38</td>\n",
              "      <td>나</td>\n",
              "      <td>38</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>선수</td>\n",
              "      <td>35</td>\n",
              "      <td>선수</td>\n",
              "      <td>36</td>\n",
              "      <td>남자</td>\n",
              "      <td>34</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>성</td>\n",
              "      <td>35</td>\n",
              "      <td>남자</td>\n",
              "      <td>34</td>\n",
              "      <td>말</td>\n",
              "      <td>30</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>남자</td>\n",
              "      <td>34</td>\n",
              "      <td>이</td>\n",
              "      <td>34</td>\n",
              "      <td>밤</td>\n",
              "      <td>26</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>중</td>\n",
              "      <td>33</td>\n",
              "      <td>그</td>\n",
              "      <td>34</td>\n",
              "      <td>낮</td>\n",
              "      <td>25</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>수</td>\n",
              "      <td>32</td>\n",
              "      <td>왜</td>\n",
              "      <td>29</td>\n",
              "      <td>뭐</td>\n",
              "      <td>25</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "  Komoran  count Twitter  count Hannanum  count\n",
              "0      강철    137      강철    137       강철    116\n",
              "1      연주    109      연주    103       연주    101\n",
              "2      !!     48       안     40        것     65\n",
              "3       말     40       거     40        수     44\n",
              "4       거     38       말     38        나     38\n",
              "5      선수     35      선수     36       남자     34\n",
              "6       성     35      남자     34        말     30\n",
              "7      남자     34       이     34        밤     26\n",
              "8       중     33       그     34        낮     25\n",
              "9       수     32       왜     29        뭐     25"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 50
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8GRvT8KjswLd",
        "colab_type": "code",
        "outputId": "c21f256d-6c51-4e6d-9412-af04ec6f035e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 359
        }
      },
      "source": [
        "# 태거들을 합하여 평균적인 결과물을 도출\n",
        "content_nouns_mix = pd.concat([content_nouns_ha_pd,content_nouns_tw_pd,content_nouns_ko_pd])\n",
        "content_nouns_mix_rank = content_nouns_mix.value_counts()[:30].reset_index()\n",
        "content_nouns_mix_rank.columns = ['Tagger Mix','count']\n",
        "content_nouns_mix_rank['count'] = content_nouns_mix_rank['count']//3\n",
        "content_nouns_mix_rank.head(10)"
      ],
      "execution_count": 51,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>Tagger Mix</th>\n",
              "      <th>count</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>강철</td>\n",
              "      <td>130</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>연주</td>\n",
              "      <td>104</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>말</td>\n",
              "      <td>36</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>남자</td>\n",
              "      <td>34</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>것</td>\n",
              "      <td>34</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>거</td>\n",
              "      <td>29</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>수</td>\n",
              "      <td>28</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>밤</td>\n",
              "      <td>26</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>소리</td>\n",
              "      <td>25</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>핸드폰</td>\n",
              "      <td>25</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "  Tagger Mix  count\n",
              "0         강철    130\n",
              "1         연주    104\n",
              "2          말     36\n",
              "3         남자     34\n",
              "4          것     34\n",
              "5          거     29\n",
              "6          수     28\n",
              "7          밤     26\n",
              "8         소리     25\n",
              "9        핸드폰     25"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 51
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "S2aZKBJMuS4E",
        "colab_type": "code",
        "outputId": "d5b6f403-ad56-4864-d181-8025e6235781",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 53
        }
      },
      "source": [
        "# 태깅 클래스 사용에도 불구하고 '것, 거, 수' 등 명사로 활용키 어려운 것들은 제거하자\n",
        "print(\"before remove {}\".format(len(content_nouns_mix)))\n",
        "remove_list = ['것','거','수','중','이','듯','안','뭐','앞','!!']\n",
        "for n in remove_list:\n",
        "  content_nouns_mix.drop(content_nouns_mix[content_nouns_mix.values == n].index, inplace=True)\n",
        "print(\"before remove {}\".format(len(content_nouns_mix)))"
      ],
      "execution_count": 52,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "before remove 12004\n",
            "before remove 10199\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qORcKuTrvR1N",
        "colab_type": "code",
        "outputId": "fc6c1f89-0e42-489c-bc6e-de5dc3a38546",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 359
        }
      },
      "source": [
        "# 쓸만한 명사들의 순위가 나왔다.\n",
        "content_nouns_mix_rank = content_nouns_mix.value_counts()[:30].reset_index()\n",
        "content_nouns_mix_rank.columns = ['Tagger Mix','count']\n",
        "content_nouns_mix_rank['count'] = content_nouns_mix_rank['count']//3\n",
        "content_nouns_mix_rank.head(10)"
      ],
      "execution_count": 53,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>Tagger Mix</th>\n",
              "      <th>count</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>강철</td>\n",
              "      <td>115</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>연주</td>\n",
              "      <td>92</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>말</td>\n",
              "      <td>32</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>남자</td>\n",
              "      <td>30</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>소리</td>\n",
              "      <td>23</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>밤</td>\n",
              "      <td>21</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>핸드폰</td>\n",
              "      <td>21</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>낮</td>\n",
              "      <td>21</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>아버지</td>\n",
              "      <td>20</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>선수</td>\n",
              "      <td>19</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "  Tagger Mix  count\n",
              "0         강철    115\n",
              "1         연주     92\n",
              "2          말     32\n",
              "3         남자     30\n",
              "4         소리     23\n",
              "5          밤     21\n",
              "6        핸드폰     21\n",
              "7          낮     21\n",
              "8        아버지     20\n",
              "9         선수     19"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 53
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wfFwk4qZyAPB",
        "colab_type": "text"
      },
      "source": [
        "## 워드클라우드 그리기"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IZjr3T8pkIIz",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from wordcloud import WordCloud\n",
        "import matplotlib.pyplot as plt"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gtcChu9hlJPq",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# 워드클래스 패키지에 입력으로 사용하기 위하여 단어 : 등장횟수 를 사전형식으로 변환\n",
        "content_rank_dict = {}\n",
        "for noun, cnt in content_nouns_mix_rank.values:\n",
        "  content_rank_dict[noun]=cnt"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7baJV1UYkTR2",
        "colab_type": "code",
        "outputId": "4eed9a42-987c-4c61-a009-b2506f8a0eec",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 216
        }
      },
      "source": [
        "# 한글 출력시 에러 발생\n",
        "wordcloud = WordCloud(background_color='white').generate_from_frequencies(content_rank_dict)\n",
        "fig = plt.figure()\n",
        "plt.axis('off')\n",
        "plt.imshow(wordcloud, interpolation='bilinear')"
      ],
      "execution_count": 56,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7fc384fd4748>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 56
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAV0AAAC1CAYAAAD86CzsAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n/xTzZZi3RgmXAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EGQDxqto6Xjn",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# 한글 폰트 설치\n",
        "! apt-get install fonts-nanum*"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "c7pJBq4aoSDD",
        "colab_type": "code",
        "outputId": "6a573048-f024-449d-dcc2-15cad7146f5f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 216
        }
      },
      "source": [
        "# 한글 폰트 문제 해결\n",
        "font_path = '/usr/share/fonts/truetype/nanum/NanumBarunGothic.ttf'\n",
        "wordcloud = WordCloud(font_path = font_path, background_color='white').generate_from_frequencies(content_rank_dict)\n",
        "fig = plt.figure()\n",
        "plt.axis('off')\n",
        "plt.imshow(wordcloud, interpolation='bilinear')"
      ],
      "execution_count": 58,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7fc385393b38>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 58
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAV0AAAC1CAYAAAD86CzsAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KCtywyBFsdm9",
        "colab_type": "text"
      },
      "source": [
        "## 기초통계 수행\n",
        "* 대사의 길이\n",
        "* 등장인물별 대사의 비중\n",
        "* 단어의 종류"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CfLlBeqRpv1T",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "script_df.content = script_df.content.map(lambda x: x.rstrip('\\n'))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6n9Uf878sw_J",
        "colab_type": "code",
        "outputId": "f0b574db-6251-4dad-97bc-d4047919ffa1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 419
        }
      },
      "source": [
        "script_df"
      ],
      "execution_count": 60,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>act</th>\n",
              "      <th>content</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>씬/1</td>\n",
              "      <td>아테네 올림픽 자료화면 인서트</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>지문</td>\n",
              "      <td>자막 – 2004년 8월. 아테네 올림픽\\n한국 국가대표 경기 생중계 자료화면.\\n...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>씬/2</td>\n",
              "      <td>방송 스튜디오 (과거, 밤)</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>지문</td>\n",
              "      <td>올림픽 중계 스튜디오. 아테네 올림픽 엠블럼과 마스코트를 배경으로 남자 MC가 진행...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>MC</td>\n",
              "      <td>네, 잠시 종목을 바꿔보겠습니다. 지금 사격 경기장에서는 50미터 권총 남자 결선이...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>574</th>\n",
              "      <td>지문</td>\n",
              "      <td>연주, 못 알아듣자 답답해 전화를 끊어버리는.\\n밖에서는 석범이 문 두드리며 ‘야 ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>575</th>\n",
              "      <td>연주</td>\n",
              "      <td>(살아있는 사람에게 말 걸 듯) 왜 나를 찾아 당신...? 날 찾아서 어쩌려고...?</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>576</th>\n",
              "      <td>지문</td>\n",
              "      <td>그림 속의 강철, 대답이 없고.</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>577</th>\n",
              "      <td>연주</td>\n",
              "      <td>왜 내가 당신 인생의 키가 되는데..?</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>578</th>\n",
              "      <td>지문</td>\n",
              "      <td>답도 없는 만화에 대고 묻는 연주와, 마치 연주에게 말을 거는 듯한 만화 속의 강철...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>579 rows × 2 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "     act                                            content\n",
              "0    씬/1                                  아테네 올림픽 자료화면 인서트 \n",
              "1     지문  자막 – 2004년 8월. 아테네 올림픽\\n한국 국가대표 경기 생중계 자료화면.\\n...\n",
              "2    씬/2                                    방송 스튜디오 (과거, 밤)\n",
              "3     지문  올림픽 중계 스튜디오. 아테네 올림픽 엠블럼과 마스코트를 배경으로 남자 MC가 진행...\n",
              "4     MC  네, 잠시 종목을 바꿔보겠습니다. 지금 사격 경기장에서는 50미터 권총 남자 결선이...\n",
              "..   ...                                                ...\n",
              "574   지문  연주, 못 알아듣자 답답해 전화를 끊어버리는.\\n밖에서는 석범이 문 두드리며 ‘야 ...\n",
              "575   연주    (살아있는 사람에게 말 걸 듯) 왜 나를 찾아 당신...? 날 찾아서 어쩌려고...?\n",
              "576   지문                                  그림 속의 강철, 대답이 없고.\n",
              "577   연주                              왜 내가 당신 인생의 키가 되는데..?\n",
              "578   지문  답도 없는 만화에 대고 묻는 연주와, 마치 연주에게 말을 거는 듯한 만화 속의 강철...\n",
              "\n",
              "[579 rows x 2 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 60
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "j3eg6xZXtNMi",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# 대사에만 집중하기 위해 지문, 씬, C#등 제거 - '지문', '씬/.*', 'C#.*'\n",
        "script_df_actor = script_df.drop(script_df[script_df['act'].str.find('지문') == 0].index)\n",
        "script_df_actor = script_df_actor.drop(script_df_actor[script_df_actor['act'].str.find('씬/') == 0].index)\n",
        "script_df_actor = script_df_actor.drop(script_df_actor[script_df_actor['act'].str.find('C#') == 0].index)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yczOaoeott-e",
        "colab_type": "code",
        "outputId": "c64b6467-4554-4578-d4bf-b22bbdf25dd9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 89
        }
      },
      "source": [
        "script_df_actor['act'].unique()"
      ],
      "execution_count": 62,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array(['MC', '캐스터', '해설', '가족들', '강철모', '강철부', '아이들', '앵커', '기자', '경감',\n",
              "       '전문가', '현석', '강철', '철호', '자막', '연주', '석범', '김간호사', '간호사1', '간호사2',\n",
              "       '박교수', '수봉', '선미/윤희', '성무', '연주/수봉', '선미', '  ', '일동', '웨이터',\n",
              "       '의사1', '매니저', '구급대원', '윤희', '소희', '도윤'], dtype=object)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 62
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gX4mt99zwR1q",
        "colab_type": "code",
        "outputId": "95372960-fad5-450c-a269-42e12b08bc81",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 80
        }
      },
      "source": [
        "# 전처리 누락분 제거\n",
        "script_df_actor[script_df_actor['act']=='  ']"
      ],
      "execution_count": 63,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>act</th>\n",
              "      <th>content</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>373</th>\n",
              "      <td></td>\n",
              "      <td>지글지글 고기 굽는 소리, 칼질하는 소리, 매운 연기에 라운지에서 들려오는 음악소리...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "    act                                            content\n",
              "373      지글지글 고기 굽는 소리, 칼질하는 소리, 매운 연기에 라운지에서 들려오는 음악소리..."
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 63
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ydy9aC8qw2Rp",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "script_df_actor = script_df_actor.drop(373)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DHd3eI-8x8C5",
        "colab_type": "text"
      },
      "source": [
        "### 등장인물의 종류"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mjf7EEVMxE7s",
        "colab_type": "code",
        "outputId": "e0910a23-7335-4ddf-808f-809360f1068f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 89
        }
      },
      "source": [
        "script_df_actor['act'].unique()"
      ],
      "execution_count": 65,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array(['MC', '캐스터', '해설', '가족들', '강철모', '강철부', '아이들', '앵커', '기자', '경감',\n",
              "       '전문가', '현석', '강철', '철호', '자막', '연주', '석범', '김간호사', '간호사1', '간호사2',\n",
              "       '박교수', '수봉', '선미/윤희', '성무', '연주/수봉', '선미', '일동', '웨이터', '의사1',\n",
              "       '매니저', '구급대원', '윤희', '소희', '도윤'], dtype=object)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 65
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4bwPC5WR40wl",
        "colab_type": "code",
        "outputId": "84eaf5a3-ad16-41ed-a1de-bca024e2ad7a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        }
      },
      "source": [
        "len(script_df_actor['act'].unique())"
      ],
      "execution_count": 66,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "34"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 66
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "A29TcMyvyB56",
        "colab_type": "text"
      },
      "source": [
        "### 대사의 길이"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZYf0Rr5qyUzp",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import seaborn as sns"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "iUC_N_Y6x6Of",
        "colab_type": "code",
        "outputId": "dbcffc26-d4a3-4acc-8996-25e734b2ccd3",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 297
        }
      },
      "source": [
        "# 대사 길이의 분포를 보면 50자 이내의 대사가 대부분을 차지함을 알 수 있다.\n",
        "sns.distplot(script_df_actor['content'].str.len())"
      ],
      "execution_count": 68,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7fc3850feac8>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 68
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYAAAAEGCAYAAABsLkJ6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nABARyVAKABGRDKUAEBHJUP8fgQPElUu8EfoAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nYY6bU9yyFek",
        "colab_type": "code",
        "outputId": "4a50673b-b16c-4e81-b2d1-c78a43a6bdd8",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 125
        }
      },
      "source": [
        "print(\"summary of content length\\n\\\n",
        "max : {}\\n\\\n",
        "min : {}\\n\\\n",
        "mean : {}\\n\\\n",
        "std : {}\\n\".format(\n",
        "            script_df_actor['content'].str.len().max(),\n",
        "            script_df_actor['content'].str.len().min(),\n",
        "            script_df_actor['content'].str.len().mean(),\n",
        "            script_df_actor['content'].str.len().std()))"
      ],
      "execution_count": 69,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "summary of content length\n",
            "max : 167\n",
            "min : 2\n",
            "mean : 33.976133651551315\n",
            "std : 29.041125819143765\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cxLFbgvg0_xH",
        "colab_type": "text"
      },
      "source": [
        "## 인물별 대사의 비중"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3UWIipq81KEj",
        "colab_type": "text"
      },
      "source": [
        "* 대사 횟수"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pqkmr0JV0-3V",
        "colab_type": "code",
        "outputId": "13d3b2aa-40ef-4574-9cf7-a71f5e180763",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 215
        }
      },
      "source": [
        "# 연주가 혼자 다해먹나보다\n",
        "script_df_actor['act'].value_counts().head(10)"
      ],
      "execution_count": 70,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "연주     139\n",
              "수봉      62\n",
              "박교수     41\n",
              "강철      28\n",
              "캐스터     25\n",
              "소희      19\n",
              "웨이터     14\n",
              "석범      13\n",
              "해설      13\n",
              "기자      13\n",
              "Name: act, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 70
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1hXvF1uL1Xoj",
        "colab_type": "text"
      },
      "source": [
        "* 대사 길이"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LeuohxyjytXZ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "script_df_actor['length'] = script_df_actor['content'].str.len()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1AK-WVEZ1bwz",
        "colab_type": "code",
        "outputId": "e44866f0-8f23-4b4a-ef7a-1b7a36ce72a4",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 390
        }
      },
      "source": [
        "# 역시 연주가 가장 말이 많다. 강철이라는 인물은 남자라 그런지 등장에 비해 대사 길이가 짧다.\n",
        "script_df_actor.groupby('act').sum().sort_values(by='length', ascending=False).head(10)"
      ],
      "execution_count": 72,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>length</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>act</th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>연주</th>\n",
              "      <td>4046</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>수봉</th>\n",
              "      <td>1853</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>박교수</th>\n",
              "      <td>1454</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>캐스터</th>\n",
              "      <td>1379</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>기자</th>\n",
              "      <td>870</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>소희</th>\n",
              "      <td>665</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>강철</th>\n",
              "      <td>633</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>해설</th>\n",
              "      <td>558</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>매니저</th>\n",
              "      <td>456</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>웨이터</th>\n",
              "      <td>369</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "     length\n",
              "act        \n",
              "연주     4046\n",
              "수봉     1853\n",
              "박교수    1454\n",
              "캐스터    1379\n",
              "기자      870\n",
              "소희      665\n",
              "강철      633\n",
              "해설      558\n",
              "매니저     456\n",
              "웨이터     369"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 72
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zyvIJWAG3bSp",
        "colab_type": "text"
      },
      "source": [
        "### 어휘의 다양성"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OHoqy6sj1jR1",
        "colab_type": "code",
        "outputId": "5566cd53-2b58-4ada-e19f-f90eef26e81a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        }
      },
      "source": [
        "# 1편의 대본에 명자 2040개가 사용되었다.\n",
        "len(content_nouns_mix.unique())"
      ],
      "execution_count": 73,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "2040"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 73
        }
      ]
    }
  ]
}