jieba_on_my_paper.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1. 使用jieba分词，得到文本中包含的词及词出现的次数 "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Building prefix dict from the default dictionary ...\n",
      "Loading model from cache c:\\users\\txy\\appdata\\local\\temp\\jieba.cache\n",
      "Loading model cost 0.407 seconds.\n",
      "Prefix dict has been built succesfully.\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import math\n",
    "import jieba\n",
    "\n",
    "# 首先对文本进行处理，得到 my_paper.txt文件\n",
    "#   包括去除标点符号、英文字符等\n",
    "\n",
    "# encoding=utf-8\n",
    "file = open('my_paper.txt')\n",
    "all_the_text = file.read()\n",
    "\n",
    "\n",
    "# 载入自定义字典，强调单词“地震动”\n",
    "jieba.load_userdict(\"userdict.txt\")\n",
    "\n",
    "# 生成一个 list\n",
    "words = jieba.lcut(all_the_text, cut_all=False)\n",
    "\n",
    "import collections\n",
    "# 生成一个字典，键是字，字是出现的次数\n",
    "dic = collections.Counter(words)\n",
    "\n",
    "# 出现的次数进行排序，频次：values\n",
    "values = sorted(dic.values(),reverse=True)\n",
    "\n",
    "# 将次数从高到低对应的键进行排序\n",
    "keys = sorted(dic, key=dic.__getitem__, reverse=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.查看高频词"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "的 956\n",
      "地震动 441\n",
      "结构 385\n",
      "持时 219\n",
      "强度 194\n",
      "和 165\n",
      "在 160\n",
      "为 136\n",
      "参数 133\n",
      "网壳 126\n",
      "对 112\n",
      "模型 109\n",
      "地震 109\n",
      "响应 108\n",
      "与 98\n",
      "研究 95\n",
      "单层 90\n",
      "中 87\n",
      "分析 87\n",
      "计算 85\n",
      "了 84\n",
      "损伤 81\n",
      "时 81\n",
      "作用 80\n",
      "下 77\n",
      "是 74\n",
      "进行 72\n",
      "频率 71\n",
      "球面 70\n",
      "动力 67\n"
     ]
    }
   ],
   "source": [
    "# 看一下出现频率高的都是什么词\n",
    "# 根据结果调整一些词的词频\n",
    "for i in range(30):\n",
    "    print keys[i], values[i]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3.词序与词频关系可视化"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# 统计一共有多少个词\n",
    "length_of_words = len(keys)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# 次序编号 1-1838\n",
    "\n",
    "# x(n)\n",
    "x = np.arange(1, length_of_words+1)\n",
    "\n",
    "# log_x(log_n)\n",
    "log_x = np.zeros(length_of_words)\n",
    "for i in range(length_of_words):\n",
    "    log_x[i] = math.log(x[i])\n",
    "    \n",
    "# 次出现的频率：xd_values\n",
    "sum_all = float(sum(values))\n",
    "xd_values = np.zeros(length_of_words)\n",
    "for i in range(length_of_words):\n",
    "    xd_values[i] = values[i]/sum_all\n",
    "\n",
    "# 频率的log值：log_xd_values\n",
    "log_xd_values = np.zeros(length_of_words)\n",
    "for i in range(length_of_words):\n",
    "    log_xd_values[i] = math.log(xd_values[i])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZgAAAEPCAYAAAB/WNKuAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAWNmZkk4YMzMLAkHjJmZJfH/AUlPSjlbNoFMAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xb9b51d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 画图\n",
    "# 序号 x 和 出现次数 values\n",
    "%matplotlib inline\n",
    "from matplotlib import pyplot as plt\n",
    "fig, ax = plt.subplots()\n",
    "ax.scatter(x, values);\n",
    "ax.set_xlabel('order');\n",
    "ax.set_ylabel('times');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "log_x = np.array(log_x)\n",
    "log_xd_values = np.array(log_xd_values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "斜率是 [[-1.10453629]]\n",
      "截距是 [-1.33838408]\n"
     ]
    }
   ],
   "source": [
    "# 用log_x和log_xd_values做线性回归\n",
    "from sklearn import linear_model\n",
    "model = linear_model.LinearRegression()\n",
    "model.fit(log_x.reshape(len(log_x),1), log_xd_values.reshape(len(log_xd_values),1))\n",
    "coef = model.coef_\n",
    "intercept = model.intercept_\n",
    "print '斜率是',coef\n",
    "print '截距是',intercept"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXwAAAEACAYAAACwB81wAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xe2437b8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "xx = np.linspace(-0.1,10,100)\n",
    "yy = (xx*(coef) + intercept).reshape(100,1)\n",
    "# log_x 和 log_xd_values 的散点图\n",
    "# y = - x 的斜线\n",
    "xx = np.linspace(-0.1,10,100)\n",
    "fig, ax = plt.subplots()\n",
    "ax.scatter(log_x, log_xd_values);\n",
    "ax.plot(xx, xx*(-1));\n",
    "ax.plot(xx, yy);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "### 4.结论"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "由齐夫定律的内容可知，判定一个文本中词是否符合齐夫定律取决于其取对数后 词序与词频之间是否符合斜率为-1的线性关系。由上述分析结果可知，该分本分析的结果是-1.1045，基本满足齐夫定律。"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 5. 词数分析"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 基于全部出现于文本中的词的分析结果表明，线性回归斜率为-1.1045，与-1还有一定差距。\n",
    "- 本节研究若取前 n 个高频单词进行分析，会不会得出什么不同的结果。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# 首先将上述过程做成函数\n",
    "# n：只分析前n个高频单词\n",
    "def words_num(n):\n",
    "    log_xx = log_x[:n]\n",
    "    log_xd_valuess = log_xd_values[:n]\n",
    "    model = linear_model.LinearRegression()\n",
    "    model.fit(log_xx.reshape(len(log_xx),1), log_xd_valuess.reshape(len(log_xd_valuess),1))\n",
    "    coef = model.coef_\n",
    "    intercept = model.intercept_\n",
    "    return coef"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "coefs = np.zeros(length_of_words)\n",
    "# 从 5 开始，计算词数量与斜率的关系\n",
    "for i in range(5, length_of_words):\n",
    "    coefs[i] = words_num(i)[0][0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA4sAAADXCAYAAACzt8RyAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1085a4a8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(15, 3))\n",
    "ax.plot(x[4:-1], coefs[5:]);\n",
    "ax.plot(x, np.zeros_like(x)-1, lw=0.5, color=\"black\");\n",
    "ax.set_xlabel('order');\n",
    "ax.set_ylabel('coef');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[-0.98208972 -0.94810726 -0.93680532 -0.9134144  -0.89128274]\n"
     ]
    }
   ],
   "source": [
    "print coefs[6:11]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "由上图可知，该文本对齐夫定律的满足程度随取得单词的数量的变化而变化。\n",
    "- 当单词数很少时（5）个词，非常满足齐夫定律（斜率为-0.98），但是数量太少了（2-4个）也不行。\n",
    "- 随着选取单词数量的增加，文本对齐夫定律的满足程度呈：“减弱——增强——减弱”的趋势，在单词数800附近又一次达到了完全满足定律的情况"
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [Root]",
   "language": "python",
   "name": "Python [Root]"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}