数据分析/机器学习/2020年02月19日/结巴分词.ipynb · 1264644959/qqq

加入 Gitee

与超过 1200万开发者一起发现、参与优秀开源项目，私有仓库也完全免费：）

该仓库未声明开源许可证文件（LICENSE），使用请关注具体项目描述及其代码上游依赖。

克隆/下载

结巴分词.ipynb 245.66 KB

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 结巴分词使用\n",
    "---\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 一、安装\n",
    "终端输入\n",
    "```pip install jieba\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "import jieba"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\"hello world, I'm  dancer\"\n",
    "我渐渐的喜欢上了你"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 二、分词模式\n",
    "---\n",
    "结巴中文分词支持的三种分词模式包括：\n",
    "\n",
    "(1) 精确模式：试图将句子最精确地切开，适合文本分析；\n",
    "\n",
    "(2) 全模式：把句子中所有的可以成词的词语都扫描出来, 速度非常快，但是不能解决歧义问题；\n",
    "\n",
    "(3) 搜索引擎模式：在精确模式的基础上，对长词再次切分，提高召回率，适合用于搜索引擎分词。"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "分词方法：\n",
    "   - jieba.cut()  生成迭代器对象\n",
    "   - jieba.lcut() 生成列表对象\n",
    "   - jieba.cut_for_search() 搜索引擎模式\n",
    "\n",
    "参数：\n",
    "   - cut_all=True 全模式\n",
    "   - cut_all=False 精确模式"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "text = \"桃花坞里桃花庵， 桃花坞里桃花仙，桃花仙人种桃树，再买桃花换酒钱\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "精确模式： 桃花坞/里/桃花庵/，/ /桃花坞/里/桃花仙/，/桃花/仙人/种/桃树/，/再/买/桃花/换/酒钱\n",
      "全模式： 桃花/桃花坞/花坞/里/桃花/桃花庵/，/ //桃花/桃花坞/花坞/里/桃花/桃花仙/，/桃花/桃花仙/仙人/人种/桃树/，/再/买/桃花/换/酒钱\n",
      "搜索引擎模式： 桃花/花坞/桃花坞/里/桃花/桃花庵/，/ /桃花/花坞/桃花坞/里/桃花/桃花仙/，/桃花/仙人/种/桃树/，/再/买/桃花/换/酒钱\n"
     ]
    }
   ],
   "source": [
    "print(\"精确模式：\",'/'.join(jieba.lcut(text, cut_all=False)))\n",
    "\n",
    "print(\"全模式：\",'/'.join(jieba.lcut(text, cut_all=True)))\n",
    "\n",
    "print(\"搜索引擎模式：\",'/'.join(jieba.lcut_for_search(text)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "text_samples = [\"桃花坞里桃花庵\",\n",
    "                \"桃花坞里桃花仙\",\n",
    "                \"桃花仙人种桃树\",\n",
    "                \"再买桃花换酒钱\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "cv = CountVectorizer()\n",
    "data = cv.fit_transform(text_samples)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['再买桃花换酒钱', '桃花仙人种桃树', '桃花坞里桃花仙', '桃花坞里桃花庵']"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cv.get_feature_names()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 现在使用jieba分词，来对文本集重新拆解\n",
    "split_text_samples = []\n",
    "for text in text_samples:\n",
    "    split_text = \" \".join(jieba.lcut(text))\n",
    "    split_text_samples.append(split_text)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "桃花坞 里 桃花庵\n",
      "桃花坞 里 桃花仙\n",
      "桃花 仙人 种 桃树\n",
      "再 买 桃花 换 酒钱\n"
     ]
    }
   ],
   "source": [
    "for text in split_text_samples:\n",
    "    print(text)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>仙人</th>\n",
       "      <th>桃树</th>\n",
       "      <th>桃花</th>\n",
       "      <th>桃花仙</th>\n",
       "      <th>桃花坞</th>\n",
       "      <th>桃花庵</th>\n",
       "      <th>酒钱</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   仙人  桃树  桃花  桃花仙  桃花坞  桃花庵  酒钱\n",
       "0   0   0   0    0    1    1   0\n",
       "1   0   0   0    1    1    0   0\n",
       "2   1   1   1    0    0    0   0\n",
       "3   0   0   1    0    0    0   1"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cv = CountVectorizer()\n",
    "data = cv.fit_transform(split_text_samples)\n",
    "DataFrame(data=data.toarray(), columns=cv.get_feature_names())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 三、新词识别\n",
    "\n",
    "jieba分词可以识别一些词典中不存在的词。但并不一定都会识别出来。比如“蛤跟”\n",
    "```\n",
    "text = \"在京教中心，没有什么女朋友是一顿烧烤哄不好的。\"\n",
    "jieba.lcut(text)\n",
    "```\n",
    "\n",
    "```['在', '京教', '中心', '，', '没有', '什么', '女朋友', '是', '一顿', '烧烤', '哄', '不好', '的', '。']```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 四、自定义词典\n",
    "\n",
    "如“美丽的多瑙河畔”,会被识别为“多瑙河”,“多瑙河畔”，“河畔”多个词汇，这是很多分词工具的共同缺陷，为此，jieba分词支持开发者使用自定义的词典，以便更精准的识别一些自创词汇。"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "添加自定义词典：\n",
    "\n",
    "    1. 添加一个.txt格式的文件，必须以utf-8编码\n",
    "    2. 文件内一个词占一行，分为三部分：词语 词频 词性(可省略，ns为地点名词)\n",
    "        eg. 多瑙河畔 1 n\n",
    "    3. jieba.load_userdict('mydict.txt')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": [
    "text = \"美丽的互助村\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['美丽', '的', '互助', '村']"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "jieba.lcut(text)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "# txt必须是一个utf-8编码\n",
    "# 词  词频  词性\n",
    "jieba.load_userdict('worddict.txt')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['美丽', '的', '互助村']"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "jieba.lcut(text)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 五、关键词提取"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "```python\n",
    "import jieba.analyse\n",
    "jieba.analyse.extract_tags(sentence, topK)\n",
    "```\n",
    "\n",
    "- sentence 待提取的文本\n",
    "- topK 返回几个TF-IDF权重最大的关键词，默认值为20个\n",
    "    \n",
    "     - TF: 词频，即某个词在文档中出现的次数\n",
    "     - IDF：逆文档频率，在词频相同的基础上，常见的词分配较小的权重，不常见的词分配较大的权重，这个权重就是逆文档频率。\n",
    "    \n",
    "     - TF-IDF：就是TF和IDF的乘积，这个值越大，说明对应的词越重要\n",
    "     \n",
    "     - [TF-IDF](http://www.ruanyifeng.com/blog/2013/03/tf-idf.html)推荐文章"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.feature_extraction.text import TfidfTransformer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open('叶问.txt', 'r',encoding='utf-8') as fp:\n",
    "    message = fp.read()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'\\ufeff到此为止吧，已经是为赋新词强说愁。模式照搬叶问2，情节粗放+强行收尾，编不下去了一场打斗全都解决了，港片通病。又把个人强弱上升到阶级层面，但观点很割裂，一方面练武是要与啦啦队同台表演，另一方面还要代表民族分出胜负，这不是闹呢，感觉这是编导的高级黑。动作场面依然是最大看点，叶伟信调动观众情绪的功力十足，让几场对决热血贲张，观感起码不错。PS，李小龙的角色是客串，但在故事层面上是错位的，依照情节安排，最终一战应该小龙哥上才对\\n\\n珍惜吧，以后拍功夫片的人越来越少\\n\\n出乎意料的好看，特别是打戏，拳拳到肉，很过瘾。 剧情也比预期的好，返璞归真了，没有太夸张的剧情。按照之前的套路，本来以为这集会越编越离谱的。 父子线的感情戏也处理得挺好，最后一场电话戏还挺感人的。 结尾小彩蛋（对前三部的闪回）同样加分不少。 最后，叶问毫无疑问就是甄子丹最好的角色。\\n\\n全片没什么尿点，相对前面三部激情多了很多，打斗不断，开场直接开始燃，一直高潮不断，李小龙跳出来的时候后面赛博朋克的色调让我眼前一亮，最后在说出自己有癌症的时候让我泪目，煽情时间不长，不脱节奏，整部片很饱满，感谢一代宗师。\\n\\n许《好莱坞往事》美国人打李小龙，就不许《叶问4》叶问去打美国人吗？？ 豆瓣er是怎么了？ “叶问是练武之人，遇到不公义的事情，一定要站出来。” 这句台词真让人听了血热。\\n\\n先说喜欢的四场打戏。1、受伤的叶问用揪耳、掐鼻、拍屁股三招稳如泰山地教育不良少年，很前辈，很大家；2、李小龙右手前伸控制敌人攻击距离，左手从右腋下向身后顺势接住对方抡空的双截棍，右手此刻再化拳弹开对手，酷炫夺棍达成；3、叶问和万师父比武，走的是《精武英雄》风的击打回合制（《功夫之王》也是），分不出胜负，突然旧金山发生大地震，两大武林宗师狼狈躲进桌下，切磋结束，都是凡人而已；4、终极对决，叶问左臂、肋骨均受重创，看上去他快被打死了，不得不用败形脚连续踢对方要害，解围后用标指刺喉，天呐，不可能再有宗师功夫片这么拍了。虽然角色古怪逻辑（吴建豪）、民族主义脸谱和《叶问2》套路都是扣分项，但，对我而言，真的，及格了。\\n\\n比想象中好很多，是一个还不错的收尾，虽不足1 2但比3要好，想起当初08年第一次看叶问的时候去到广州旅游还特意要去佛山看一看咏春拳，一晃丹哥老了，系列结束了，我也老了。'"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "message"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['叶问', '李小龙', '打戏', '功夫片', '对决', '角色', '打斗', '宗师', '练武', '收尾']"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import jieba.analyse\n",
    "jieba.analyse.extract_tags(message, 10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 如果自己去提取重要词汇，可以拿到词权重"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'\\ufeff 到此为止 吧 ， 已经 是 为 赋 新词 强说 愁 。 模式 照搬 叶问 2 ， 情节 粗放 + 强行 收尾 ， 编不 下去 了 一场 打斗 全都 解决 了 ， 港片 通病 。 又 把 个人 强弱 上升 到 阶级 层面 ， 但 观点 很 割裂 ， 一方面 练武 是 要 与 啦啦队 同台 表演 ， 另一方面 还要 代表 民族 分出 胜负 ， 这 不是 闹 呢 ， 感觉 这是 编导 的 高级 黑 。 动作 场面 依然 是 最大 看点 ， 叶伟信 调动 观众 情绪 的 功力 十足 ， 让 几场 对决 热血 贲张 ， 观感 起码 不错 。 PS ， 李小龙 的 角色 是 客串 ， 但 在 故事 层面 上 是 错位 的 ， 依照 情节 安排 ， 最终 一战 应该 小龙 哥上 才 对 \\n \\n 珍惜 吧 ， 以后 拍 功夫片 的 人 越来越少 \\n \\n 出乎意料 的 好看 ， 特别 是 打戏 ， 拳拳 到 肉 ， 很 过瘾 。   剧情 也 比 预期 的 好 ， 返璞归真 了 ， 没有 太 夸张 的 剧情 。 按照 之前 的 套路 ， 本来 以为 这 集会 越 编越 离谱 的 。   父子 线 的 感情 戏 也 处理 得 挺 好 ， 最后 一场 电话 戏 还 挺 感人 的 。   结尾 小 彩蛋 （ 对前 三部 的 闪回 ） 同样 加 分 不少 。   最后 ， 叶问 毫无疑问 就是 甄子丹 最好 的 角色 。 \\n \\n 全片 没什么 尿点 ， 相对 前面 三部 激情 多 了 很多 ， 打斗 不断 ， 开场 直接 开始 燃 ， 一直 高潮 不断 ， 李小龙 跳 出来 的 时候 后面 赛博 朋克 的 色调 让 我 眼前一亮 ， 最后 在 说出 自己 有 癌症 的 时候 让 我 泪目 ， 煽情 时间 不 长 ， 不 脱 节奏 ， 整部 片 很 饱满 ， 感谢 一代宗师 。 \\n \\n 许 《 好莱坞 往事 》 美国 人 打 李小龙 ， 就 不许 《 叶问 4 》 叶问 去 打 美国 人 吗 ？ ？   豆瓣 er 是 怎么 了 ？   “ 叶问 是 练武 之 人 ， 遇到 不 公义 的 事情 ， 一定 要 站 出来 。 ”   这句 台词 真让人 听 了 血热 。 \\n \\n 先说 喜欢 的 四场 打戏 。 1 、 受伤 的 叶 问用 揪 耳 、 掐 鼻 、 拍 屁股 三招 稳如泰山 地 教育 不良少年 ， 很 前辈 ， 很 大家 ； 2 、 李小龙 右手 前伸 控制 敌人 攻击 距离 ， 左手 从右 腋下 向 身后 顺势 接 住 对方 抡 空 的 双截棍 ， 右手 此刻 再化拳 弹开 对手 ， 酷炫夺 棍 达成 ； 3 、 叶问 和 万 师父 比武 ， 走 的 是 《 精武 英雄 》 风 的 击打 回合制 （ 《 功夫 之王 》 也 是 ） ， 分不出 胜负 ， 突然 旧金山 发生 大 地震 ， 两大 武林 宗师 狼狈 躲进 桌下 ， 切磋 结束 ， 都 是 凡人 而已 ； 4 、 终极 对决 ， 叶问 左臂 、 肋骨 均 受 重创 ， 看上去 他 快 被 打死 了 ， 不得不 用败 形脚 连续 踢 对方 要害 ， 解围 后用 标指 刺喉 ， 天呐 ， 不 可能 再有 宗师 功夫片 这么 拍 了 。 虽然 角色 古怪 逻辑 （ 吴建豪 ） 、 民族主义 脸谱 和 《 叶问 2 》 套路 都 是 扣分 项 ， 但 ， 对 我 而言 ， 真的 ， 及格 了 。 \\n \\n 比 想象 中好 很多 ， 是 一个 还 不错 的 收尾 ， 虽 不足 1   2 但 比 3 要 好 ， 想起 当初 08 年 第一次 看叶 问 的 时候 去 到 广州 旅游 还 特意 要 去 佛山 看一看 咏春拳 ， 一晃 丹哥 老 了 ， 系列 结束 了 ， 我 也 老 了 。'"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "content = \" \".join(jieba.lcut(message))\n",
    "content"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [],
   "source": [
    "stop_words = ['\\ufeff','，','吧','是','。','+','了','又',\"\\n\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [],
   "source": [
    "good_message = \"\"\n",
    "for word in jieba.cut(message):\n",
    "    if word not in stop_words:\n",
    "        good_message = good_message + word"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'到此为止已经为赋新词强说愁模式照搬叶问2情节粗放强行收尾编不下去一场打斗全都解决港片通病把个人强弱上升到阶级层面但观点很割裂一方面练武要与啦啦队同台表演另一方面还要代表民族分出胜负这不是闹呢感觉这是编导的高级黑动作场面依然最大看点叶伟信调动观众情绪的功力十足让几场对决热血贲张观感起码不错PS李小龙的角色客串但在故事层面上错位的依照情节安排最终一战应该小龙哥上才对珍惜以后拍功夫片的人越来越少出乎意料的好看特别打戏拳拳到肉很过瘾 剧情也比预期的好返璞归真没有太夸张的剧情按照之前的套路本来以为这集会越编越离谱的 父子线的感情戏也处理得挺好最后一场电话戏还挺感人的 结尾小彩蛋（对前三部的闪回）同样加分不少 最后叶问毫无疑问就是甄子丹最好的角色全片没什么尿点相对前面三部激情多很多打斗不断开场直接开始燃一直高潮不断李小龙跳出来的时候后面赛博朋克的色调让我眼前一亮最后在说出自己有癌症的时候让我泪目煽情时间不长不脱节奏整部片很饱满感谢一代宗师许《好莱坞往事》美国人打李小龙就不许《叶问4》叶问去打美国人吗？？ 豆瓣er怎么？ “叶问练武之人遇到不公义的事情一定要站出来” 这句台词真让人听血热先说喜欢的四场打戏1、受伤的叶问用揪耳、掐鼻、拍屁股三招稳如泰山地教育不良少年很前辈很大家；2、李小龙右手前伸控制敌人攻击距离左手从右腋下向身后顺势接住对方抡空的双截棍右手此刻再化拳弹开对手酷炫夺棍达成；3、叶问和万师父比武走的《精武英雄》风的击打回合制（《功夫之王》也）分不出胜负突然旧金山发生大地震两大武林宗师狼狈躲进桌下切磋结束都凡人而已；4、终极对决叶问左臂、肋骨均受重创看上去他快被打死不得不用败形脚连续踢对方要害解围后用标指刺喉天呐不可能再有宗师功夫片这么拍虽然角色古怪逻辑（吴建豪）、民族主义脸谱和《叶问2》套路都扣分项但对我而言真的及格比想象中好很多一个还不错的收尾虽不足1 2但比3要好想起当初08年第一次看叶问的时候去到广州旅游还特意要去佛山看一看咏春拳一晃丹哥老系列结束我也老'"
      ]
     },
     "execution_count": 97,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "good_message"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [],
   "source": [
    "content = \" \".join(jieba.lcut(good_message))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'到此为止 已经 为 赋 新词 强说 愁 模式 照搬 叶问 2 情节 粗放 强行 收尾 编不 下去 一场 打斗 全都 解决 港片 通病 把 个人 强弱 上升 到 阶级 层面 但 观点 很 割裂 一方面 练武 要 与 啦啦队 同台 表演 另一方面 还要 代表 民族 分出 胜负 这 不是 闹 呢 感觉 这是 编导 的 高级 黑 动作 场面 依然 最大 看点 叶伟信 调动 观众 情绪 的 功力 十足 让 几场 对决 热血 贲张 观感 起码 不错 PS 李小龙 的 角色 客串 但 在 故事 层面 上 错位 的 依照 情节 安排 最终 一战 应该 小龙 哥上 才 对 珍惜 以后 拍 功夫片 的 人 越来越少 出乎意料 的 好看 特别 打戏 拳拳 到 肉 很 过瘾   剧情 也 比 预期 的 好 返璞归真 没有 太 夸张 的 剧情 按照 之前 的 套路 本来 以为 这 集会 越 编越 离谱 的   父子 线 的 感情 戏 也 处理 得 挺 好 最后 一场 电话 戏 还 挺 感人 的   结尾 小 彩蛋 （ 对前 三部 的 闪回 ） 同样 加 分 不少   最后 叶问 毫无疑问 就是 甄子丹 最好 的 角色 全片 没什么 尿点 相对 前面 三部 激情 多 很多 打斗 不断 开场 直接 开始 燃 一直 高潮 不断 李小龙 跳 出来 的 时候 后面 赛博 朋克 的 色调 让 我 眼前一亮 最后 在 说出 自己 有 癌症 的 时候 让 我 泪目 煽情 时间 不长 不脱 节奏 整部 片 很 饱满 感谢 一代宗师 许 《 好莱坞 往事 》 美国 人 打 李小龙 就 不许 《 叶问 4 》 叶问 去 打 美国 人 吗 ？ ？   豆瓣 er 怎么 ？   “ 叶问 练武 之 人 遇到 不 公义 的 事情 一定 要 站 出来 ”   这句 台词 真让人 听血 热先 说 喜欢 的 四场 打戏 1 、 受伤 的 叶 问用 揪 耳 、 掐 鼻 、 拍 屁股 三招 稳如泰山 地 教育 不良少年 很 前辈 很 大家 ； 2 、 李小龙 右手 前伸 控制 敌人 攻击 距离 左手 从右 腋下 向 身后 顺势 接 住 对方 抡 空 的 双截棍 右手 此刻 再化拳 弹开 对手 酷炫夺 棍 达成 ； 3 、 叶问 和 万 师父 比武 走 的 《 精武 英雄 》 风 的 击打 回合制 （ 《 功夫 之王 》 也 ） 分不出 胜负 突然 旧金山 发生 大 地震 两大 武林 宗师 狼狈 躲进 桌下 切磋 结束 都 凡人 而已 ； 4 、 终极 对决 叶问 左臂 、 肋骨 均 受 重创 看上去 他 快 被 打死 不得不 用败 形脚 连续 踢 对方 要害 解围 后用 标指 刺 喉天 呐 不 可能 再有 宗师 功夫片 这么 拍 虽然 角色 古怪 逻辑 （ 吴建豪 ） 、 民族主义 脸谱 和 《 叶问 2 》 套路 都 扣分 项 但 对 我 而言 真的 及格 比 想象 中好 很多 一个 还 不错 的 收尾 虽 不足 1   2 但 比 3 要 好 想起 当初 08 年 第一次 看叶 问 的 时候 去 到 广州 旅游 还 特意 要 去 佛山 看一看 咏春拳 一晃 丹哥 老 系列 结束 我 也 老'"
      ]
     },
     "execution_count": 99,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "content"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [],
   "source": [
    "tfv = TfidfVectorizer()\n",
    "data = tfv.fit_transform([content])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.09656091, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.09656091, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.09656091, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.09656091, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.09656091, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.09656091, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.09656091, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.09656091, 0.04828045, 0.38624364, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.09656091, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.09656091, 0.04828045, 0.09656091, 0.04828045,\n",
       "        0.04828045, 0.09656091, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.09656091, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.09656091, 0.04828045,\n",
       "        0.04828045, 0.09656091, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.09656091, 0.09656091,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.09656091, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.14484136,\n",
       "        0.04828045, 0.14484136, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.19312182, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.09656091, 0.04828045, 0.04828045, 0.09656091,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.09656091, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.09656091, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.14484136, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045, 0.04828045, 0.04828045, 0.04828045, 0.04828045,\n",
       "        0.04828045]])"
      ]
     },
     "execution_count": 103,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.toarray()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['08',\n",
       " 'er',\n",
       " 'ps',\n",
       " '一个',\n",
       " '一代宗师',\n",
       " '一场',\n",
       " '一定',\n",
       " '一战',\n",
       " '一方面',\n",
       " '一晃',\n",
       " '一直',\n",
       " '三招',\n",
       " '三部',\n",
       " '上升',\n",
       " '下去',\n",
       " '不少',\n",
       " '不得不',\n",
       " '不断',\n",
       " '不是',\n",
       " '不脱',\n",
       " '不良少年',\n",
       " '不许',\n",
       " '不足',\n",
       " '不错',\n",
       " '不长',\n",
       " '两大',\n",
       " '个人',\n",
       " '中好',\n",
       " '丹哥',\n",
       " '之前',\n",
       " '之王',\n",
       " '事情',\n",
       " '从右',\n",
       " '代表',\n",
       " '以为',\n",
       " '以后',\n",
       " '佛山',\n",
       " '依然',\n",
       " '依照',\n",
       " '全片',\n",
       " '全都',\n",
       " '公义',\n",
       " '再化拳',\n",
       " '再有',\n",
       " '几场',\n",
       " '凡人',\n",
       " '出乎意料',\n",
       " '出来',\n",
       " '击打',\n",
       " '分不出',\n",
       " '分出',\n",
       " '切磋',\n",
       " '到此为止',\n",
       " '前伸',\n",
       " '前辈',\n",
       " '前面',\n",
       " '剧情',\n",
       " '割裂',\n",
       " '功力',\n",
       " '功夫',\n",
       " '功夫片',\n",
       " '动作',\n",
       " '十足',\n",
       " '及格',\n",
       " '双截棍',\n",
       " '发生',\n",
       " '受伤',\n",
       " '古怪',\n",
       " '另一方面',\n",
       " '可能',\n",
       " '台词',\n",
       " '右手',\n",
       " '叶伟信',\n",
       " '叶问',\n",
       " '同台',\n",
       " '同样',\n",
       " '后用',\n",
       " '后面',\n",
       " '听血',\n",
       " '吴建豪',\n",
       " '咏春拳',\n",
       " '哥上',\n",
       " '啦啦队',\n",
       " '喉天',\n",
       " '喜欢',\n",
       " '四场',\n",
       " '回合制',\n",
       " '地震',\n",
       " '场面',\n",
       " '处理',\n",
       " '大家',\n",
       " '夸张',\n",
       " '套路',\n",
       " '好看',\n",
       " '好莱坞',\n",
       " '安排',\n",
       " '宗师',\n",
       " '客串',\n",
       " '对决',\n",
       " '对前',\n",
       " '对手',\n",
       " '对方',\n",
       " '小龙',\n",
       " '就是',\n",
       " '尿点',\n",
       " '屁股',\n",
       " '层面',\n",
       " '左手',\n",
       " '左臂',\n",
       " '已经',\n",
       " '师父',\n",
       " '广州',\n",
       " '应该',\n",
       " '开场',\n",
       " '开始',\n",
       " '弹开',\n",
       " '强弱',\n",
       " '强行',\n",
       " '强说',\n",
       " '当初',\n",
       " '形脚',\n",
       " '彩蛋',\n",
       " '往事',\n",
       " '很多',\n",
       " '怎么',\n",
       " '情绪',\n",
       " '情节',\n",
       " '想象',\n",
       " '想起',\n",
       " '感人',\n",
       " '感情',\n",
       " '感觉',\n",
       " '感谢',\n",
       " '打戏',\n",
       " '打斗',\n",
       " '打死',\n",
       " '扣分',\n",
       " '拳拳',\n",
       " '按照',\n",
       " '控制',\n",
       " '收尾',\n",
       " '攻击',\n",
       " '故事',\n",
       " '敌人',\n",
       " '教育',\n",
       " '整部',\n",
       " '新词',\n",
       " '旅游',\n",
       " '旧金山',\n",
       " '时候',\n",
       " '时间',\n",
       " '最后',\n",
       " '最大',\n",
       " '最好',\n",
       " '最终',\n",
       " '朋克',\n",
       " '本来',\n",
       " '李小龙',\n",
       " '标指',\n",
       " '桌下',\n",
       " '模式',\n",
       " '此刻',\n",
       " '武林',\n",
       " '比武',\n",
       " '毫无疑问',\n",
       " '民族',\n",
       " '民族主义',\n",
       " '没什么',\n",
       " '没有',\n",
       " '泪目',\n",
       " '港片',\n",
       " '激情',\n",
       " '热先',\n",
       " '热血',\n",
       " '照搬',\n",
       " '煽情',\n",
       " '父子',\n",
       " '特别',\n",
       " '特意',\n",
       " '狼狈',\n",
       " '珍惜',\n",
       " '甄子丹',\n",
       " '用败',\n",
       " '电话',\n",
       " '癌症',\n",
       " '直接',\n",
       " '相对',\n",
       " '看一看',\n",
       " '看上去',\n",
       " '看叶',\n",
       " '看点',\n",
       " '真的',\n",
       " '真让人',\n",
       " '眼前一亮',\n",
       " '离谱',\n",
       " '稳如泰山',\n",
       " '突然',\n",
       " '第一次',\n",
       " '粗放',\n",
       " '精武',\n",
       " '系列',\n",
       " '练武',\n",
       " '终极',\n",
       " '结尾',\n",
       " '结束',\n",
       " '编不',\n",
       " '编导',\n",
       " '编越',\n",
       " '美国',\n",
       " '而已',\n",
       " '而言',\n",
       " '肋骨',\n",
       " '胜负',\n",
       " '脸谱',\n",
       " '腋下',\n",
       " '自己',\n",
       " '色调',\n",
       " '节奏',\n",
       " '英雄',\n",
       " '虽然',\n",
       " '表演',\n",
       " '要害',\n",
       " '观众',\n",
       " '观感',\n",
       " '观点',\n",
       " '角色',\n",
       " '解决',\n",
       " '解围',\n",
       " '说出',\n",
       " '调动',\n",
       " '豆瓣',\n",
       " '贲张',\n",
       " '赛博',\n",
       " '起码',\n",
       " '越来越少',\n",
       " '距离',\n",
       " '身后',\n",
       " '躲进',\n",
       " '达成',\n",
       " '过瘾',\n",
       " '返璞归真',\n",
       " '还要',\n",
       " '这么',\n",
       " '这句',\n",
       " '这是',\n",
       " '连续',\n",
       " '通病',\n",
       " '逻辑',\n",
       " '遇到',\n",
       " '酷炫夺',\n",
       " '重创',\n",
       " '错位',\n",
       " '闪回',\n",
       " '问用',\n",
       " '阶级',\n",
       " '集会',\n",
       " '顺势',\n",
       " '预期',\n",
       " '饱满',\n",
       " '高潮',\n",
       " '高级']"
      ]
     },
     "execution_count": 104,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tfv.get_feature_names()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [],
   "source": [
    "importance_words = DataFrame(data=data.toarray(), columns=tfv.get_feature_names()).loc[0].sort_values(ascending=False)[:50]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "叶问     0.386244\n",
       "李小龙    0.193122\n",
       "时候     0.144841\n",
       "最后     0.144841\n",
       "角色     0.144841\n",
       "对方     0.096561\n",
       "结束     0.096561\n",
       "层面     0.096561\n",
       "不错     0.096561\n",
       "美国     0.096561\n",
       "对决     0.096561\n",
       "收尾     0.096561\n",
       "不断     0.096561\n",
       "剧情     0.096561\n",
       "练武     0.096561\n",
       "出来     0.096561\n",
       "三部     0.096561\n",
       "功夫片    0.096561\n",
       "打斗     0.096561\n",
       "打戏     0.096561\n",
       "很多     0.096561\n",
       "胜负     0.096561\n",
       "一场     0.096561\n",
       "套路     0.096561\n",
       "右手     0.096561\n",
       "情节     0.096561\n",
       "宗师     0.096561\n",
       "回合制    0.048280\n",
       "地震     0.048280\n",
       "四场     0.048280\n",
       "高级     0.048280\n",
       "喜欢     0.048280\n",
       "喉天     0.048280\n",
       "啦啦队    0.048280\n",
       "哥上     0.048280\n",
       "处理     0.048280\n",
       "咏春拳    0.048280\n",
       "吴建豪    0.048280\n",
       "听血     0.048280\n",
       "后面     0.048280\n",
       "后用     0.048280\n",
       "同样     0.048280\n",
       "同台     0.048280\n",
       "叶伟信    0.048280\n",
       "场面     0.048280\n",
       "客串     0.048280\n",
       "大家     0.048280\n",
       "夸张     0.048280\n",
       "彩蛋     0.048280\n",
       "形脚     0.048280\n",
       "Name: 0, dtype: float64"
      ]
     },
     "execution_count": 123,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "importance_words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 词云  重要词汇的展示方式"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 六、去除停用词\n",
    "\n",
    "在信息检索中，为节省存储空间和提高搜索效率，在处理自然语言数据（或文本）之前或之后会自动过滤掉某些字或词，比如“的”、“是”、“而且”、“但是”、”非常“等。这些字或词即被称为Stop Words（停用词）。"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 七、使用sklearn计算TF-IDF值"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "```\n",
    "corpus = ['aaa ccc aaa aaa', \n",
    "          'aaa aaa', \n",
    "          'aaa aaa aaa', \n",
    "          'aaa aaa aaa aaa',\n",
    "          'aaa bbb aaa bbb aaa',\n",
    "          'ccc aaa aaa ccc aaa'\n",
    "         ]\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "corpus = ['aaa ccc aaa aaa', \n",
    "          'aaa aaa', \n",
    "          'aaa aaa aaa', \n",
    "          'aaa aaa aaa aaa',\n",
    "          'aaa bbb aaa bbb aaa',\n",
    "          'ccc aaa aaa ccc aaa'\n",
    "         ]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.feature_extraction.text import CountVectorizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "cv = CountVectorizer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 有监督学习\n",
    "knn.fit(X, y)\n",
    "knn.predict(X)\n",
    "knn.score()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 无监督学习\n",
    "modle.fit(X)\n",
    "model.predict(X)\n",
    "\n",
    "model.fit_predict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = cv.fit_transform(corpus)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<6x3 sparse matrix of type '<class 'numpy.int64'>'\n",
       "\twith 9 stored elements in Compressed Sparse Row format>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "samples \n",
    "{X1,    {x11,x12,x13...}\n",
    " X2,    {x21,x22,x23}\n",
    " X3,\n",
    " X4}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['aaa ccc aaa aaa',\n",
       " 'aaa aaa',\n",
       " 'aaa aaa aaa',\n",
       " 'aaa aaa aaa aaa',\n",
       " 'aaa bbb aaa bbb aaa',\n",
       " 'ccc aaa aaa ccc aaa']"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "corpus"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['aaa', 'bbb', 'ccc']"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cv.get_feature_names()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# data是一个matric矩阵，可以通过toarray()函数，提取其中的信息\n",
    "train = data.toarray()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pandas import Series, DataFrame"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>aaa</th>\n",
       "      <th>bbb</th>\n",
       "      <th>ccc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   aaa  bbb  ccc\n",
       "0    3    0    1\n",
       "1    2    0    0\n",
       "2    3    0    0\n",
       "3    4    0    0\n",
       "4    3    2    0\n",
       "5    3    0    2"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "DataFrame(data=train, columns=cv.get_feature_names())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "词频向量的转换过程\n",
    "\n",
    "1. 构造一个词频向量转换对象\n",
    "    from sklearn.feature_extraction.text import CountVectorizer\n",
    "    cv = CountVectorizer()\n",
    "2. 使用cv对象转换待转换的文本集合\n",
    "    注意：文本集是一个一维数组\n",
    "    data = cv.fit_transform(text_samples)\n",
    "3. 获取转换结果\n",
    "    获取特征： cv.get_featrue_names()【注意从cv中获取而不是data】\n",
    "    获取转换的向量集：data.toarray()\n",
    "    \n",
    "4. 如果只转换成词频向量，无法更精准的表达语句的侧重点，所以要结合逆文档来进行转换\n",
    "    from sklearn.feature_extraction.text import TfidfTransformer\n",
    "    tf = TfidfTransformer()\n",
    "    tf_data = tf.fit_transform(data)\n",
    "    \n",
    "5. 可以把两次的转换综合成一次转换\n",
    "    from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "    \n",
    "    tfv = TfidfVectorizer()\n",
    "    data = tfv.fit_transform(text_samples)\n",
    "    tfv.get_feature_names()\n",
    "    data.toarray()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "hello toney , where are you??"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "我喜欢你，不是因为你有房有车，而是在那个阳光明媚的午后，你恰好穿了一件白衬衫！"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "今天心情好，今天去吃饭，后来心情不好了"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "千锋学习，我上个月也在学习，但是没在千锋  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.feature_extraction.text import TfidfVectorizer, TfidfTransformer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 现在使用逆文档频率对词频向量再次做转换处理\n",
    "tf = TfidfTransformer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "tf_train = tf.fit_transform(train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[3, 0, 1],\n",
       "       [2, 0, 0],\n",
       "       [3, 0, 0],\n",
       "       [4, 0, 0],\n",
       "       [3, 2, 0],\n",
       "       [3, 0, 2]], dtype=int64)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.85151335, 0.        , 0.52433293],\n",
       "       [1.        , 0.        , 0.        ],\n",
       "       [1.        , 0.        , 0.        ],\n",
       "       [1.        , 0.        , 0.        ],\n",
       "       [0.55422893, 0.83236428, 0.        ],\n",
       "       [0.63035731, 0.        , 0.77630514]])"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# TF-IDF转换后的最终结果\n",
    "# 这里考虑了 词频大小 单词的逆文档频率\n",
    "tf_train.toarray()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 直接转换成TF-IDF的向量表示\n",
    "tfv = TfidfVectorizer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = tfv.fit_transform(corpus)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.85151335, 0.        , 0.52433293],\n",
       "       [1.        , 0.        , 0.        ],\n",
       "       [1.        , 0.        , 0.        ],\n",
       "       [1.        , 0.        , 0.        ],\n",
       "       [0.55422893, 0.83236428, 0.        ],\n",
       "       [0.63035731, 0.        , 0.77630514]])"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.toarray()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['aaa', 'bbb', 'ccc']"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tfv.get_feature_names()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>aaa</th>\n",
       "      <th>bbb</th>\n",
       "      <th>ccc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.851513</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.524333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.554229</td>\n",
       "      <td>0.832364</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.630357</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.776305</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        aaa       bbb       ccc\n",
       "0  0.851513  0.000000  0.524333\n",
       "1  1.000000  0.000000  0.000000\n",
       "2  1.000000  0.000000  0.000000\n",
       "3  1.000000  0.000000  0.000000\n",
       "4  0.554229  0.832364  0.000000\n",
       "5  0.630357  0.000000  0.776305"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "DataFrame(data=data.toarray(), columns=tfv.get_feature_names())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 练习 读取test.txt文件，获取词频集\n",
    "\n",
    "```\n",
    "#读取预料 一行预料为一个文档 \n",
    "for line in open('test.txt', 'r').readlines():\n",
    "    #print line\n",
    "    corpus.append(line.strip())\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 生成一组汉语的文本集\n",
    "\n",
    "text = ['天之道',\n",
    "        '损有余而补不足',\n",
    "        '是故虚胜实',\n",
    "        '不足胜有余',\n",
    "        '其意博',\n",
    "        '其理奥',\n",
    "        '其趣深',\n",
    "        '天地之象分',\n",
    "        '阴阳之候列'\n",
    "       ]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "cv = CountVectorizer()\n",
    "data = cv.fit_transform(text)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0, 1, 0, 0, 0, 0, 0, 0, 0],\n",
       "       [0, 0, 0, 0, 1, 0, 0, 0, 0],\n",
       "       [0, 0, 0, 0, 0, 1, 0, 0, 0],\n",
       "       [1, 0, 0, 0, 0, 0, 0, 0, 0],\n",
       "       [0, 0, 0, 1, 0, 0, 0, 0, 0],\n",
       "       [0, 0, 0, 0, 0, 0, 1, 0, 0],\n",
       "       [0, 0, 0, 0, 0, 0, 0, 1, 0],\n",
       "       [0, 0, 1, 0, 0, 0, 0, 0, 0],\n",
       "       [0, 0, 0, 0, 0, 0, 0, 0, 1]], dtype=int64)"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.toarray()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['不足胜有余', '天之道', '天地之象分', '意博', '损有余而补不足', '是故虚胜实', '理奥', '趣深', '阴阳之候列']"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 因为默认的词频转换，是以空格作为拆分标志的，而英文具有得天独厚的优势，但是汉语不行，所以需要借助jieba分词来处理\n",
    "cv.get_feature_names()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>不足胜有余</th>\n",
       "      <th>天之道</th>\n",
       "      <th>天地之象分</th>\n",
       "      <th>意博</th>\n",
       "      <th>损有余而补不足</th>\n",
       "      <th>是故虚胜实</th>\n",
       "      <th>理奥</th>\n",
       "      <th>趣深</th>\n",
       "      <th>阴阳之候列</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   不足胜有余  天之道  天地之象分  意博  损有余而补不足  是故虚胜实  理奥  趣深  阴阳之候列\n",
       "0      0    1      0   0        0      0   0   0      0\n",
       "1      0    0      0   0        1      0   0   0      0\n",
       "2      0    0      0   0        0      1   0   0      0\n",
       "3      1    0      0   0        0      0   0   0      0\n",
       "4      0    0      0   1        0      0   0   0      0\n",
       "5      0    0      0   0        0      0   1   0      0\n",
       "6      0    0      0   0        0      0   0   1      0\n",
       "7      0    0      1   0        0      0   0   0      0\n",
       "8      0    0      0   0        0      0   0   0      1"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "DataFrame(data=data.toarray(), columns=cv.get_feature_names())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 词云的展示\n",
    "1. 统计词频 （Collection.Counter对象）(结合TfidfVectorizer确定停用词， CountVectoreizer统计词频)\n",
    "2. 预加载词云的展示格式 \n",
    "    ```\n",
    "    wc = wordcloun.WordClound(font_path=\"自定义字体的路径\", \n",
    "                      max_words=30,     # 字体的个数的上限\n",
    "                      max_font_size=100,  # 字体的最大字号\n",
    "                      min_font_size=10,\n",
    "                      mask=\"设置背景图片\")\n",
    "    ```\n",
    "3. 加载资源   \n",
    "    - wc.generate_from_frequencies(词频的统计量Counter对象)  # 加载文字资源\n",
    "    - wc.recolor(color_func=image_colors)     # 加载颜色资源\n",
    "    \n",
    "    其中，image_colors的获取如下：\n",
    "    \n",
    "        #解析背景图片颜色\n",
    "        - image_colors = wordcloud.ImageColorGenerator(plt.imread('图片路径'))\n",
    "4. plt.imshow() "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [],
   "source": [
    "import wordcloud"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "叶问     0.386244\n",
       "李小龙    0.193122\n",
       "时候     0.144841\n",
       "最后     0.144841\n",
       "角色     0.144841\n",
       "对方     0.096561\n",
       "结束     0.096561\n",
       "层面     0.096561\n",
       "不错     0.096561\n",
       "美国     0.096561\n",
       "对决     0.096561\n",
       "收尾     0.096561\n",
       "不断     0.096561\n",
       "剧情     0.096561\n",
       "练武     0.096561\n",
       "出来     0.096561\n",
       "三部     0.096561\n",
       "功夫片    0.096561\n",
       "打斗     0.096561\n",
       "打戏     0.096561\n",
       "很多     0.096561\n",
       "胜负     0.096561\n",
       "一场     0.096561\n",
       "套路     0.096561\n",
       "右手     0.096561\n",
       "情节     0.096561\n",
       "宗师     0.096561\n",
       "回合制    0.048280\n",
       "地震     0.048280\n",
       "四场     0.048280\n",
       "高级     0.048280\n",
       "喜欢     0.048280\n",
       "喉天     0.048280\n",
       "啦啦队    0.048280\n",
       "哥上     0.048280\n",
       "处理     0.048280\n",
       "咏春拳    0.048280\n",
       "吴建豪    0.048280\n",
       "听血     0.048280\n",
       "后面     0.048280\n",
       "后用     0.048280\n",
       "同样     0.048280\n",
       "同台     0.048280\n",
       "叶伟信    0.048280\n",
       "场面     0.048280\n",
       "客串     0.048280\n",
       "大家     0.048280\n",
       "夸张     0.048280\n",
       "彩蛋     0.048280\n",
       "形脚     0.048280\n",
       "Name: 0, dtype: float64"
      ]
     },
     "execution_count": 124,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# s 描述词和重要性的一种列表结构\n",
    "importance_words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x1c809a02448>"
      ]
     },
     "execution_count": 119,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(plt.imread('yewen4.jpg'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [],
   "source": [
    "back_image = plt.imread('yewen4.jpg')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x1c80c28df88>"
      ]
     },
     "execution_count": 129,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x864 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 词云图片的基本配置\n",
    "# mask=backimage\n",
    "wc = wordcloud.WordCloud(font_path=\"FZZJ-LongYTJW.TTF\", \n",
    "                   max_words=60,     # 字体的个数的上限\n",
    "                   max_font_size=100,  # 字体的最大字号\n",
    "                   min_font_size=1,\n",
    "                         mask=back_image)\n",
    "# 绘制文字 \n",
    "image = wc.generate_from_frequencies(importance_words)\n",
    "\n",
    "image_colors = wordcloud.ImageColorGenerator(back_image)\n",
    "wc.recolor(color_func=image_colors)\n",
    "\n",
    "\n",
    "plt.figure(figsize=(12,12))\n",
    "# 展示数据\n",
    "plt.imshow(image)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 使用词频向量来展示数据的重要程度"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "朴素贝叶斯"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.3857142857142857"
      ]
     },
     "execution_count": 133,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(0.9*(3/7))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0002777777777777778"
      ]
     },
     "execution_count": 134,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(2/(360*20))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0005833333333333334"
      ]
     },
     "execution_count": 136,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(0.9*(2/(360*20))/(3/7))"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}

一键复制编辑原始数据按行查看历史

提交于 2020-02-19 18:03 . 提交

https://gitee.com/super__man/qqq.git

git@gitee.com:super__man/qqq.git

super__man

qqq

master

1264644959 / qqq

简介

发行版

贡献者

近期动态

1264644959 / qqq .gitee-modal { width: 500px !important; }

简介

发行版

贡献者

近期动态

搜索帮助

1264644959 / qqq