NLTK/NLTK01-文本数据集操作.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# NLTK自带图书数据集"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "*** Introductory Examples for the NLTK Book ***\n",
      "Loading text1, ..., text9 and sent1, ..., sent9\n",
      "Type the name of the text or sentence to view it.\n",
      "Type: 'texts()' or 'sents()' to list the materials.\n",
      "text1: Moby Dick by Herman Melville 1851\n",
      "text2: Sense and Sensibility by Jane Austen 1811\n",
      "text3: The Book of Genesis\n",
      "text4: Inaugural Address Corpus\n",
      "text5: Chat Corpus\n",
      "text6: Monty Python and the Holy Grail\n",
      "text7: Wall Street Journal\n",
      "text8: Personals Corpus\n",
      "text9: The Man Who Was Thursday by G . K . Chesterton 1908\n"
     ]
    }
   ],
   "source": [
    "from nltk.book import *\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 查看文本信息"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Text: Sense and Sensibility by Jane Austen 1811>"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "text2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Text: Moby Dick by Herman Melville 1851>"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "text1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 在文本中搜索"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Displaying 25 of 49 matches:\n",
      "des cometh within the chaos of this monster ' s mouth , be it beast , boat , or\n",
      "nter into the dreadful gulf of this monster ' s ( whale ' s ) mouth , are immed\n",
      "time with a lance ; but the furious monster at length rushed on the boat ; hims\n",
      " . Such a portentous and mysterious monster roused all my curiosity . Then the \n",
      "and flank with the most exasperated monster . Long usage had , for this Stubb ,\n",
      "ACK ).-- Under this head I reckon a monster which , by the various names of Fin\n",
      "arned the history of that murderous monster against whom I and all the others h\n",
      "ocity , cunning , and malice in the monster attacked ; therefore it was , that \n",
      "iathan is restricted to the ignoble monster primitively pursued in the North ; \n",
      " and incontestable character of the monster to strike the imagination with unwo\n",
      "mberment . Then , in darting at the monster , knife in hand , he had but given \n",
      "e rock ; instead of this we saw the monster sailing off with the utmost gravity\n",
      "e at Constantinople , a great sea - monster was captured in the neighboring Pro\n",
      " Of what precise species this sea - monster was , is not mentioned . But as he \n",
      "man reasoning , Procopius ' s sea - monster , that for half a century stove the\n",
      "hale ,\" as he called the fictitious monster which he declared to be incessantly\n",
      "d his intention to hunt that mortal monster in person . But such a supposition \n",
      "ng us on and on , in order that the monster might turn round upon us , and rend\n",
      "d famous , and most deadly immortal monster , Don ;-- but that would be too lon\n",
      "oluntarily lifted his voice for the monster , though for some little time past \n",
      "s rescuing Andromeda from the sea - monster or whale . Where did Guido get the \n",
      " huge corpulence of that Hogarthian monster undulates on the surface , scarcely\n",
      "nd is drawn just balancing upon the monster ' s spine ; and standing in that pr\n",
      " of cutting - in ) hove over to the monster as if to a quay ; and a boat , hurr\n",
      "eet in length . They fancy that the monster to which these arms belonged ordina\n"
     ]
    }
   ],
   "source": [
    "text1.concordance('monster')    # 根据词索引"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 查看词的上下文"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "whale ship world sea whales boat pequod other sun leviathan thing king\n",
      "water head captain air crew cabin body more\n"
     ]
    }
   ],
   "source": [
    "text1.similar('monster')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "and fool curate\n"
     ]
    }
   ],
   "source": [
    "text9.similar('monster')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 查看多个词共同上下文\n",
    "\n",
    "我们使用一个词列表作为参数"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "a_pretty am_glad a_lucky is_pretty be_glad\n"
     ]
    }
   ],
   "source": [
    "text2.common_contexts(['monstrous', 'very'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 词频率散点图\n",
    "\n",
    "可以研究一段文本中，随着时间推移，词语使用的变化情况"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZsAAAEWCAYAAACwtjr+AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nO3deZxcVZ338c83BGgxmrDkUVSSBkZRFMyYRoEB0zi4ISK+BgZ4dCQOTMRRx+gwGidqGkcFlFFxGQF9tJXBhXWGQR3g0adFwoB0EAgoOxFZlB0J+/J7/rjnUjc3VdXV3XV6Md/361WvunXOuef87qlb9ctdUq2IwMzMLKcZkx2AmZn96XOyMTOz7JxszMwsOycbMzPLzsnGzMyyc7IxM7PsnGxsgyXpJ5IOHWcfiyVdOM4+rpbUP54+uqkb8zKGMQck/ftEjmkTy8nGpgVJayTt3c0+I+LNEfGdbvZZJalXUkhamx5/kHSOpNfX4nh5RAzlimO0cs2LpEFJj6e5uFfS+ZJeOoZ+ur4vWH5ONmb5zYmIWcArgfOBsyQtnqxgJM2crLGBz6W5eBFwJzA4ibHYBHKysWlP0r6SLpd0v6SLJO2cyrdP/4J+VXr9Akl3laesJA1JOrzSz99J+o2kByX9urLeMkk3VsrfPpY4I+L3EXE8MAAcK2lG6v+Zf6lLerWkYUl/TEdCX0jl5VHSEkm3S7pD0pGV2GdU4rxH0qmStqite5ikW4CfSeqR9O+p7f2SLpX0vPq8pH4/Lum3ku6U9F1Js2v9HirpFkl3S1re4Vw8DHwPeEWzekn7pdOL96d4XpbKTwbmAf+VjpA+Mtr3wSaHk41Na5L+HPgW8B5gS+BE4GxJm0bEjcBHgX+XtBnwbeA7zU5ZSTqQIgm8C3gusB9wT6q+EdgTmA0clfrbehxhnwn8L2CHJnXHA8dHxHOB7YFTa/V7AS8G3gB8tHI66QPA/sAi4AXAfcDXausuAl4GvBE4NG3PNhTzdgTwSJN4FqfHXsB2wCzgq7U2e6Rt+Uvgk2ViaEfSLOAdwK+a1L0E+D6wFJgL/JgiuWwSEX8D3AK8NSJmRcTnRhrLpgYnG5vulgAnRsQlEfFUutbwGLArQER8A7gBuATYGmj1L+/DKU7xXBqFGyLit6mP0yLi9oh4OiJ+CFwPvHocMd+enrdoUvcE8GeStoqItRFxca3+qIh4KCJWUyTPQ1L5EcDyiLg1Ih6jSJwH1E6ZDaR1H0njbAn8WZq3VRHxxybxvAP4QkTcFBFrgY8BB9f6PSoiHomIK4ArKE4XtnKkpPsp3pNZFIms7iDgRxFxfkQ8ARwHPAvYvU2/NsU52dh0Nx/4x3S65f70RbYNxb/uS9+gOF3zlfRF3Mw2FEcw65H0rsppuvtTX1uNI+YXpud7m9QdBrwEuCad2tq3Vv+7yvJvaWznfIprQWWMvwGeAp7XYt2TgXOBH6TTcp+TtHGTeF6QxqmOObPW7+8ryw9TJJFWjouIORHx/IjYLx19th0zIp5Osb+wSVubJpxsbLr7HfCZ9AVWPjaLiO/DM6drvgT8H2CgvI7Rop/t64WS5lMkq/cDW0bEHOAqQOOI+e0UF8evrVdExPURcQjFabZjgdMlPbvSZJvK8jwaR0m/A95cm4eeiLit2n1lnCci4qiI2JHiiGFfilOIdbdTJLLqmE8Cf+hwW8dinTEliWK7y23xT9VPQ042Np1snC5sl4+ZFIngCEmvUeHZkt4i6TlpneOB4Yg4HPgRcEKLvr9JcYpnYernz1KieTbFl9tdAJLeTYuL2iOR9DxJ7wdWAB9L/2Kvt3mnpLmp7v5UXG33CUmbSXo58G7gh6n8BOAzKWYkzZX0tjax7CVpJ0kbAX+kOK22XjwU104+JGnblLg/C/wwIp4czbaP0qnAWyT9ZTra+keKU6MXpfo/UFw/smnEycamkx9TXMQuHwMRMQz8HcVF6/sorgUsBkhftm8C3pvW/zDwKknvqHccEacBn6G4Q+pB4D+ALSLi18C/Av9D8SW3E7BylHHfL+khYDWwD3BgRHyrRds3AVdLWkuRKA9O11hKP0/b+FOKU1LnpfLjgbOB8yQ9CFwMvKZNTM8HTqdINL9J/Z7cpN23UvkFwM3AoxQ3I2QTEdcC7wS+AtwNvJXihoDHU5OjgY+nU4ZHtujGphj5j6eZTX2Seim+7DfOfFRhloWPbMzMLDsnGzMzy86n0czMLDsf2ZiZWXaT+YN8U9ZWW20Vvb29kx2Gmdm0smrVqrsjYm6zOiebJnp7exkeHp7sMMzMphVJv21V59NoZmaWnZONmZll52RjZmbZOdmYmVl2TjZmZpadk42ZmWXnZGNmZtk52ZiZWXZONmZmlp2TjZmZZedkY2Zm2TnZmJlZdk42ZmaWnZONmZll52RjZmbZOdmYmVl2TjZmZpadk42ZmWXnZGNmZtk52ZiZWXZONmZmlp2TjZmZZedkY2Zm2TnZmJlZdk42ZmaWnZONmZll52RjZmbZOdmYmVl2TjZmZpadk42ZmWXnZGNmZtlNWrKROELiXWl5scQLKnXflNhxsmIzM7PumrRkE8EJEXw3vVwMjWQTweER/HpSAuuC3l4YGCge0FgeGID+/sZy2a63tygv1y2Xy3XLdarrNntdHbdeXvZVtqmrxlsdv922NFu3Wd+l/n7o6YE5c4pH2XbOnKKujLPaZ39/YzvLPuqxVNepP9dja9ZXsz7q6m3KuFqNU21fnf9y3Xrf9XXq5a3mvfp+djIfddU5r7dpFg+sH3+r9ZvF1MyMGc3HbrfeSPtwVf3zNFJ/PT2djdVKs/Hq+1x132nWd7P3slyn3Peqn42enkb78j1t9T7V95fyO6fd+9oNioi8I5QDFUcxRwIBXAncCKwF1gCDwG3AI8BuwE9S2xcAn0pdPAvYJIJtJRYCXwBmAXcDiyO4Q2IIuATYC5gDHBbBLyReDnwb2IQiwf5VBNe3irWvry+Gh4fHs63PiFj3dTvVtuXb0um6o+m72n+pOq7UfPxWfbRat67ZtozUZ6u2reaqGsNIsbcat1n8o5mTdnHVx2vWX7PyqpHiaDcfzd73+pit5qJdP+36q29Xq3WazXGn47Rbp902ddKmk3kbqa92+2Sr7ahuS7P9vpmR3u96TM3WHw9JqyKir1ndhBzZpC/7jwOvi+CVwAfLughOB4aBd0SwIIJHKnVnp7IFwBXAcRIbA18BDohgIfAt4DOV4WZG8GpgKbAilR0BHJ/66QNuzbWtZma2vpkTNM7rgNMiuBsggntH8y92iY8Aj0TwNYlXAK8Azk99bATcUWl+ZnpeBfSm5f8Blku8CDiz2VGNpCXAEoB58+Z1HpyZmY1oyt+NJrE3cCDF0QmAgKvLI54IdorgDZVVHkvPT5GSaQTfA/ajOE33Y4nX1ceJiJMioi8i+ubOnZtrc8zMNkgTlWx+BhwosSWAxBa1+geB59RXkpgPfA04sHJ67VpgrsRuqc3G6TRdSxLbATdF8GXgP4Gdx7MxZmY2OhNyGi2CqyU+A/xc4ingVxQ3BpQGgROkZ24QKC0GtgT+I50yuz2CfSQOAL4sMZtiG74EXN0mhL8G/kbiCeD3wGe7sFktzZ8Pixc3Xq9Y0VgeGmrc9TE4WLQbHGzcQTJ//rp3CK1Y0Vinum69r2p/dYODjb5atanGuGhRZ9vSrKxZfbXfiy9u3O2zdGnxPHs2LFgAa9as38/QUPFcbueiRY3l+ljl6/pztb46Z9W+Roq/vo1lXO3iKM2fv255dd1Ox2xVX30/Rzsf1diajdmqv+r+0Symdq+bKU+pd7oftapvtU413k7metNNxz5Wq/Gqc1d+lkfTd3Wfq89/fz8ccwwsW1a8Lt/TVu9TPabq909OE3Y32nQy3rvRzMw2RJN+N5qZmW3YnGzMzCw7JxszM8vOycbMzLJzsjEzs+ycbMzMLDsnGzMzy87JxszMsnOyMTOz7JxszMwsOycbMzPLzsnGzMyyc7IxM7PsnGzMzCw7JxszM8vOycbMzLJzsjEzs+ycbMzMLDsnGzMzy87JxszMsnOyMTOz7JxszMwsOycbMzPLzsnGzMyyc7IxM7PsnGzMzCw7JxszM8vOycbMzLIbdbKRGJA4MkcwNvEGBrrbx2j7GxjoTgw2OQYGoL+/u+9hq74GBqCnp7HPDAzAjPQNVo2huk/19+eJpVv6+9ePsbd33W1s9qjHNprP4Hg+r+OhiBjdCmIAWBvBcVkiGnn8mRE8mXOMvr6+GB4ezjnElCHBKHeBtn2Mtj+peB5vDDY5yvcPuvcettqHqmNVRay7H9WXxxNXNz4fI/UP647Rajur6ts2ms/geD6vI5G0KiL6mtV1dGQjsVziOokLgR1S2fYS/y2xSuIXEi9N5YMSX5e4WOImiX6Jb0n8RmKw0uchEqslrpI4tlL+JonLJK6Q+GkqG5A4WWIlcLJEbxrzsvTYvbL+R1O/V0gck+K8rFL/4uprMzPLb+ZIDSQWAgcDC1L7y4BVwEnAERFcL/Ea4N+A16XVNgd2A/YDzgb+AjgcuFRiAXAncCywELgPOE9if2Al8A3gtRHcLLFFJZQdgT0ieERiM+D1ETwq8WLg+0CfxJuBtwGvieBhiS0iuFfiAYkFEVwOvBv49vrbqSXAEoB58+Z1NntmZtaREZMNsCdwVgQPA0icDfQAuwOnVQ75Nq2s818RhMRq4A8RrE7rXg30AvOBoQjuSuWnAK8FngIuiOBmgAjurfR5dgSPpOWNga+mxPUU8JJUvjfw7TLWyvrfBN4t8WHgIODV9Y2MiJMoEih9fX0+qWNm1kWdJJtmZgD3R7CgRf1j6fnpynL5eibwxBjGfKiy/CHgD8ArUyyPjrDuGcAK4GfAqgjuGcP4ZmY2Rp0kmwuAQYmjU/u3AicCN0scGMFpEgJ2juCKDsf9JfBlia0oTqMdAnwFuBj4N4lty9NotaOb0mzg1gieljgU2CiVnw98UuKU6mm0dLrtXODrwGEdxrhBWLGiu32Mtr9ujG+TZ8UKGBoa/11f9T5blR9zDCxb1ij71KeK50WLGjFU11+0KE8s3dIsvvnzYfHikddt9bkbKebxfF7Ho6O70SSWA4dSXGu5heK6zRkUX95bU5zW+kEEn0o3AZwTwekSvWn5Famfat0hwD8DAn4UwUdTmzcDn6U4YrkzgtfX74BL12nOAAL4b+B9EcxKdcuAdwGPAz+O4J9T+a7A6cD8CJ5qt70b0t1oZmbd0u5utFHf+jxdpf8bNDuCT4zU1snGzGz02iWbsV6zmVYkzgK2p3G3nJmZTaANItlE8PbJjsHMbEPm30YzM7PsnGzMzCw7JxszM8vOycbMzLJzsjEzs+ycbMzMLDsnGzMzy87JxszMsnOyMTOz7JxszMwsOycbMzPLzsnGzMyyc7IxM7PsnGzMzCw7JxszM8vOycbMzLJzsjEzs+ycbMzMLDsnGzMzy87JxszMsnOyMTOz7JxszMwsOycbMzPLzsnGzMyyc7IxM7PssiUbiX+Q+I3EKV3ud0DiyG72OZl6e2FgoPG6v7+xXC3vpoGB9R/1+m6NUz5Xt6vVWNX2o4mh1XZMtHL8OXPWL2vXvtPy0bTr7W3ebizxVNX3z/p7W+2jOg/N4qi/Z2PZ9/v7G4+yv97exuuennXH6e8v4irXKduW7epxlMvVfbMsr++v1bmojjlnzrqxleU9PY0Yqupz0+rzWfZV7XvmzOLR07PuNjWb53KdZmPkoIjI07G4Btg7glsrZTMjeHKc/Q4AayM4bpwhttTX1xfDw8O5ul+HVDyXb4PUfDnHmFXVcbo1btlPfRubtem0fatxSpl2547jKOPv5D1sVdfp/Hfa93jjaddvqZOxxrNuu3g6Vd23RmrTbJ+sP7fqc6T6ZuX1eWoVTz2mTrTbplZjjJWkVRHR16wuy5GNxAnAdsBPJB6QOFliJXCyxEYSn5e4VOJKifdU1vunSvlRlfLlEtdJXAjsUClfIHFxan+WxOapfEjiixLD6ehqF4kzJa6X+HSObTYzs9ayJJsIjgBuB/YCvgjsSHGUcwhwGPBABLsAuwB/J7GtxBuAFwOvBhYACyVeK7EQODiV7ZPWKX0X+GgEOwOrgRWVuscj6ANOAP4TeB/wCmCxxJb1mCUtkTQsafiuu+7q2lyYmRnMnKBxzo7gkbT8BmBniQPS69kUSeYN6fGrVD4rlT8HOCuChwEkzk7Ps4E5Efw8tf8OcFp1zPS8Grg6gjvSejcB2wD3VAOMiJOAk6A4jTbeDTYzs4aJSjYPVZYFfCCCc6sNJN4IHB3BibXypWMc87H0/HRluXw9UdttZmZMzpfuucB7JX4WwRMSLwFuS+X/InFKBGslXgg8AVwADEocneJ9K3BiBA9I3CexZwS/AP4GnjnKmTbmz4fFixuvFy1qLK9YsV7zrhip326NW/azYgUMDY08VrX9WMaZbGUcs2evX9aufaflo2k3f37zdmOJp6rZ/ll9b6t9VOehkzjGsu9X1ynvshocbNxldswxsGzZuu0vvxwWLCher1lTtL344ka7ap/lcn3fXLSoMV411nIu6vOwdGkjtvLzvummsOuuRQxV9W1v9br+3TE4CLem27Fmpm/2cpuqfVS3aXCw+Rg55LwbbQ3QB7yfyt1jEjOAT1MkDQF3Afun5PFB4PDUxVrgnRHcKLEcOBS4E7gFuCyC4yQWUFyT2Qy4CXh3BPdJDAFHRjAs0Z+W903jP1PXKvaJvBvNzOxPRbu70bIlm+nMycbMbPQm/NZnMzOzKicbMzPLzsnGzMyyc7IxM7PsnGzMzCw7JxszM8vOycbMzLJzsjEzs+ycbMzMLDsnGzMzy87JxszMsnOyMTOz7JxszMwsOycbMzPLzsnGzMyyc7IxM7PsnGzMzCw7JxszM8vOycbMzLJzsjEzs+ycbMzMLDsnGzMzy87JxszMsnOyMTOz7JxszMwsOycbMzPLzsnGzMyym3LJRmJA4sg29Qsk9qm83k9i2cRE15mBgfZ1ZX39GaC/v9Gm2hagt7eobzVes3GbjdXb23yd3t5GDNV4RtJJm3o8zcrK8evmzFk/3p6e9eenWTztxmvXpl3dSP2P1Ee5fnWbyve9nU7HrfZf9t3s9Ug62baxrDOWfrs5/p/i2NXxO9k3JoMiYvJGb0JiAFgbwXEt6hcDfRG8P1cMfX19MTw8POb1JWg1rVLxHNFoV21f1lfV6+p9N+unVV21/2Zjtyob6/Z20rZd/GV9Pd76drQao914ncQ/0vpjmZ+R3pNO+ukk5mbvdSfjtIq7E93eX0YrZ99Teezq+KPdn7sbg1ZFRF+zuilxZCOxXOI6iQuBHVLZkERfWt5KYo3EJsCngIMkLpc4SGKxxFdTu7kSZ0hcmh5/kcoXpfaXS/xK4jmTtKlmZhukmZMdgMRC4GBgAUU8lwGrmrWN4HGJT1I5sklHOqXjgS9GcKHEPOBc4GXAkcD7IlgpMQt4dP04tARYAjBv3rwubZ2ZmcEUSDbAnsBZETwMIHH2OPraG9ixcrrguSm5rAS+IHEKcGYEt9ZXjIiTgJOgOI02jhjMzKxmKiSbVp6kcZqvp8N1ZgC7Rqx35HKMxI+AfYCVEm+M4JouxWlmZiOYCsnmAmBQ4miKeN4KnAisARYCvwQOqLR/EFpeczkP+ADweSjuXIvgcontI1gNrJbYBXgp5Es2K1Z0VlcuV8sWLWp9x8j8+c3v2GrWT6u6FStgcBAWL15/nfnzGzFU4xlJJ23q8TQrK8evmz0bli5dt+2mm8KyFvcgVuNpN167Nu3qRup/pD7K9avvydDQyHcKdTpuvf+hoeavR9LJto1lnbH0283x/xTHro7fyb4xGabE3WgSy4FDgTuBWyiu25wDnAo8BfwIeGcEvRJbUFyL2Rg4GngW6RqOxFbA1yiu08wELojgCImvAHsBTwNXA4sjeKxVPOO9G83MbEPU7m60KZFsphonGzOz0Zvytz6bmdmfNicbMzPLzsnGzMyyc7IxM7PsnGzMzCw7JxszM8vOycbMzLJzsjEzs+ycbMzMLDsnGzMzy87JxszMsnOyMTOz7JxszMwsOycbMzPLzsnGzMyyc7IxM7PsnGzMzCw7JxszM8vOycbMzLJzsjEzs+ycbMzMLDsnGzMzy87JxszMsnOyMTOz7JxszMwsOycbMzPLzsnGzMyyy5psJPaXCImXZuq/T+LLOfo2M7PuyX1kcwhwYXruKomZEQxH8A/d7rub+vthYKBYHhgoXudWjldf7ma/09VEb8N0mbM5c0bXvr8fenuLx8BA49Hb26gr28H689Dfv/5noVrWbt7qdXPmjBx/u/HLWFvFUY+1vx9mzGhs+8yZxfjlHJRtquOWY5T9VZ/rj/7+Rp/lc9m2nN+enqK8p6cRR38/SI32PT2N2Ov9l/2UZWU/M2cW25aDIiJPx2IWcC2wF/BfEewg0Q8cBdwP7AScCqwGPgg8C9g/ghsl5gInAPNSd0sjWCkxAGwPbAfcApwIHBnBvmm8rwB9QABHRXCGxNeBXVL/p0ewYqTY+/r6Ynh4uBvTgFQ8R6y7nJPUGKO63M1+p6uJ3obpMmejjbPcl9sp9/nqc339VmXt4umkr9GsM1JddXuq9a002+7q83Qw1n1W0qqI6GtWN3M8AY3gbcB/R3CdxD0SC1P5K4GXAfcCNwHfjODVEh8EPgAsBY4HvhjBhRLzgHPTOgA7AntE8EhKXqVPAA9EsBOAxOapfHkE90psBPxUYucIrsy21WZmtp6cyeYQiqQB8IP0+hzg0gjuAJC4ETgvtVlNcRQEsDewY+VfAc9NRy4AZ0fwSJPx9gYOLl9EcF9a/GuJJRTbujVFslov2UhaAiwBmDdvXr3azMzGIUuykdgCeB2wk0QAG1Gc2voR8Fil6dOV109X4pkB7BrBo7V+AR4aRRzbAkcCu0Rwn8Qg0NOsbUScBJwExWm0TscwM7OR5bpB4ADg5AjmR9AbwTbAzcCeHa5/HsUpNQAkFnSwzvnA+yrrbA48lyI5PSDxPODNHY5vZmZdlOs02iHAsbWyM4D3Ajd2sP4/AF+TuJIixguAI0ZY59NpnauApyhuEDhT4lfANcDvgJWdb0J3LFrUuDNlxQoYGso/5ooVzZe72e90NdHbMF3mbPbs0bVftAjWrCmWFy9ulA8OFnc1lXWLFhXP9Xkoy1uVtZu3el0nsbcbf/789nHUP7OLFsEFF8C8ecW2f/rTMGsWLF26fh/luOUYZX/9/Y3nuqEhuPDCos+1a4vnBQuKtuX8XnxxcQfZo4/C859fxDE0BD//eTEfCxYUbXbdteizOs7QUPH+lHekARxzTNHPrbfC00+vH1M3ZLsbbTrr5t1oZmYbinZ3o/kXBMzMLDsnGzMzy87JxszMsnOyMTOz7JxszMwsOycbMzPLzsnGzMyyc7IxM7PsnGzMzCw7JxszM8vOycbMzLJzsjEzs+ycbMzMLDsnGzMzy87JxszMsnOyMTOz7JxszMwsOycbMzPLzsnGzMyyc7IxM7PsnGzMzCw7JxszM8vOycbMzLJzsjEzs+ycbMzMLDsnGzMzy87JxszMsnOyMTOz7JxszMwsOycbMzPLzsnGzMyyU0RMdgxTjqS7gN+OcfWtgLu7GE4OjnH8pnp84Bi7YarHB1MrxvkRMbdZhZNNl0kajoi+yY6jHcc4flM9PnCM3TDV44PpESP4NJqZmU0AJxszM8vOyab7TprsADrgGMdvqscHjrEbpnp8MD1i9DUbMzPLz0c2ZmaWnZONmZll52TTRZLeJOlaSTdIWpZ5rG0k/T9Jv5Z0taQPpvItJJ0v6fr0vHkql6Qvp9iulPSqSl+HpvbXSzq0Ur5Q0uq0zpclaQxxbiTpV5LOSa+3lXRJ6vOHkjZJ5Zum1zek+t5KHx9L5ddKemOlfNzzLWmOpNMlXSPpN5J2m4Jz+KH0Hl8l6fuSeiZ7HiV9S9Kdkq6qlGWft1ZjjCLGz6f3+kpJZ0maM9b5Gct7MFJ8lbp/lBSStprMOeyqiPCjCw9gI+BGYDtgE+AKYMeM420NvCotPwe4DtgR+BywLJUvA45Ny/sAPwEE7Apcksq3AG5Kz5un5c1T3S9TW6V13zyGOD8MfA84J70+FTg4LZ8AvDct/z1wQlo+GPhhWt4xzeWmwLZpjjfq1nwD3wEOT8ubAHOm0hwCLwRuBp5Vmb/Fkz2PwGuBVwFXVcqyz1urMUYR4xuAmWn52EqMo56f0b4HncSXyrcBzqX4j+VbTeYcdvU7ayIG2RAewG7AuZXXHwM+NoHj/yfweuBaYOtUtjVwbVo+ETik0v7aVH8IcGKl/MRUtjVwTaV8nXYdxvQi4KfA64Bz0k5/d+XD/sycpQ/Xbml5Zmqn+jyW7box38Bsii9y1cqn0hy+EPhd+jKZmebxjVNhHoFe1v0izz5vrcboNMZa3duBU5pt90jzM5Z9udP4gNOBVwJraCSbSZvDbj18Gq17yi+F0q2pLLt0mP7nwCXA8yLijlT1e+B5I8TXrvzWJuWj8SXgI8DT6fWWwP0R8WSTPp+JI9U/kNqPNu7R2Ba4C/i2ilN935T0bKbQHEbEbcBxwC3AHRTzsoqpNY+liZi3VmOMxd9S/It/LDGOZV8ekaS3AbdFxBW1qqk6hx1zspnmJM0CzgCWRsQfq3VR/NNlUu5tl7QvcGdErJqM8Ts0k+I0xtcj4s+BhyhOKzxjMucQIJ1PfxtFYnwB8GzgTZMVT6cmYt7GM4ak5cCTwCldDWocJG0G/DPwyYkacyL3byeb7rmN4lxr6UWpLBtJG1MkmlMi4sxU/AdJW6f6rYE7R4ivXfmLmpR36i+A/SStAX5AcSrteGCOpJlN+nwmjlQ/G7hnDHGPxq3ArRFxSXp9OkXymSpzCLA3cHNE3BURTwBnUsztVJrH0kTMW6sxOiZpMbAv8I70ZTuWGO9h9O/BSLan+EfFFelz8yLgMknPH0N8WedwTCbiXN2G8KD4V/JNFDtLeSHx5RnHE/Bd4Eu18s+z7sW/z6Xlt7DuBcZfpvItKK5bbJ4eNwNbpLr6BcZ9xhhrP40bBE5j3Yuqf5+W38e6FxwR4z4AAAQ2SURBVFVPTcsvZ90LtzdRXLTtynwDvwB2SMsDaf6mzBwCrwGuBjZLfXwH+MBUmEfWv2aTfd5ajTGKGN8E/BqYW2s36vkZ7XvQSXy1ujU0rtlM2hx265F9gA3pQXHHyHUUd68szzzWHhSHv1cCl6fHPhTnhn8KXA/838qOJ+BrKbbVQF+lr78FbkiPd1fK+4Cr0jpfpcVFzg5i7aeRbLZLH4Ib0od101Tek17fkOq3q6y/PMVwLZW7ubox38ACYDjN43+kD+yUmkPgKOCa1M/JFF+IkzqPwPcpriE9QXGEeNhEzFurMUYR4w0U1zjKz8wJY52fsbwHI8VXq19DI9lMyhx28+GfqzEzs+x8zcbMzLJzsjEzs+ycbMzMLDsnGzMzy87JxszMsnOyMRsjSV+UtLTy+lxJ36y8/ldJHx5H/wOSjmxRtyT9evE1kn4paY9K3Z4qfiX6cknPSr90fLWkz49y/F5J/3us8ZtVOdmYjd1KYHcASTOArSj+c2Bpd+CiTjqq/E/0TtruC7wH2CMiXgocAXwv/U9zgHcAR0fEgoh4BFgC7BwR/9TpGEkv4GRjXeFkYzZ2F1H82i8USeYq4EFJm0vaFHgZxc+NKB1dXJX+vshBAJL6Jf1C0tkU/6sdScslXSfpQmCHFuN+FPiniLgbICIuo/hlgfdJOhz4a+BfJJ2S+p4FrJJ0kKQDUxxXSLogjblRiu/S9LdS3pPGOQbYMx0hfaibE2cbno7/NWVm64qI2yU9KWkexVHM/1D8su5uFL/0uzoiHpf0VxS/VPBKiqOfS8sveorfYntFRNwsaSHFz5ssoPhsXkbxC891L29SPgwcGhGfSKfUzomI0wEkrY2IBWl5NfDGiLhNjT8cdhjwQETskpLkSknnUfyUyZERse/4ZsrMycZsvC6iSDS7A1+gSDa7UySblanNHsD3I+Ipih9B/DmwC/BHit+4ujm12xM4KyIeBkhHJd22EhiUdCrFj3pC8QfFdpZ0QHo9G3gx8HiG8W0D5dNoZuNTXrfZieI02sUURzadXq95aAxj/hpYWCtbSPGDnW1FxBHAxyl+KXiVpC0pfnfrA+kaz4KI2DYizhtDXGYtOdmYjc9FFD9Xf29EPBUR91L8aendaCSbXwAHpWsjcyn+HPAvm/R1AbB/uoPsOcBbW4z5OeDYlCiQtIDiT0X/20jBSto+Ii6JiE9S/OG48k8Qvzf9yQokvST9EbkHKf7kuNm4+TSa2fisprgO871a2azyAj5wFkXyuYLil7o/EhG/l/TSakcRcZmkH6Z2dwKXNhswIs6W9ELgIklBkRTeGY2/vtjO5yW9mOJo5qdprCsp7jy7TJIoktD+qfwpSVcAgxHxxQ76N2vKv/psZmbZ+TSamZll52RjZmbZOdmYmVl2TjZmZpadk42ZmWXnZGNmZtk52ZiZWXb/Hx0js/xkRbZJAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "text4.dispersion_plot(['citizens', 'democracy', 'freedom', 'duties', 'America'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 根据以后文本生成随机文本\n",
    "\n",
    "生成的新文本的所有词有来自于旧文本，只是顺序变化。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Building ngram index...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "laid by her , and said unto Cain , Where art thou , and said , Go to ,\n",
      "I will not do it for ten ' s sons ; we dreamed each man according to\n",
      "their generatio the firstborn said unto Laban , Because I said , Nay ,\n",
      "but Sarah shall her name be . , duke Elah , duke Shobal , and Akan .\n",
      "and looked upon my affliction . Bashemath Ishmael ' s blood , but Isra\n",
      "for as a prince hast thou found of all the cattle in the valley , and\n",
      "the wo The\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "\"laid by her , and said unto Cain , Where art thou , and said , Go to ,\\nI will not do it for ten ' s sons ; we dreamed each man according to\\ntheir generatio the firstborn said unto Laban , Because I said , Nay ,\\nbut Sarah shall her name be . , duke Elah , duke Shobal , and Akan .\\nand looked upon my affliction . Bashemath Ishmael ' s blood , but Isra\\nfor as a prince hast thou found of all the cattle in the valley , and\\nthe wo The\""
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "text3.generate()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 查看文本的长度\n",
    "\n",
    "使用len()函数，输出文本中字符和标点的总长度。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "260819"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(text1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "使用len()函数获取到的长度是全文本长度，其中的词肯定是有重复的，如果我们想获取没有重复的词长度，换言之，我们想知道这个文本一共有哪些词组成。我们使用set()函数。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# set(text1)  # 显示内容过大，注释掉"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "19317"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(set(text1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "如果想对text的内容进行排序，我们可以用sorted()函数，它会输出一个词汇列表，从标点符号开始，然后是字母...\n",
    "\n",
    "我们前面看到文本总长度为：260819，词汇量为（包含标点）：19317，我们再来看看，平均每个词被使用了多少次"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "13.502044830977896"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(text1)/len(set(text1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "还可以看一个词在文本中出现的次数，以及在文中所占的百分比"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "49"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "text1.count('monster')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.018786974875296663"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "100 * text1.count('monster') / len(text1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "创建一个函数来进行这些计算，方便我们后续对代码的复用"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 计算文本中每个词被使用了多少次\n",
    "def lexical_diversity(text):\n",
    "    return len(text) / len(set(text))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 计算一个词在文中占据的百分比\n",
    "def percentage(count, total):\n",
    "    return 100 * count / total"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "13.502044830977896"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lexical_diversity(text1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.018786974875296663"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "percentage(text1.count('monster'), len(text1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 在Python中如何记录表达文本：列表"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "在nltk中提供了一些短句"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Call', 'me', 'Ishmael', '.']"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sent1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "['Call', 'me', 'Ishmael', '.']这种数据格式在Python中叫做`list`，列表。我们可以在这种数据上使用我们上面介绍的len()...等函数，甚至可以使用我们自己写的`lexical_diversity`和`percentage`函数"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(sent1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lexical_diversity(sent1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "list之间可以进行Python的’加法运算‘，这种加法叫做`连接`，多个list首位拼接为一个新的list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Call', 'me', 'Ishmael', '.']"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sent1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['The',\n",
       " 'family',\n",
       " 'of',\n",
       " 'Dashwood',\n",
       " 'had',\n",
       " 'long',\n",
       " 'been',\n",
       " 'settled',\n",
       " 'in',\n",
       " 'Sussex',\n",
       " '.']"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sent2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Call',\n",
       " 'me',\n",
       " 'Ishmael',\n",
       " '.',\n",
       " 'The',\n",
       " 'family',\n",
       " 'of',\n",
       " 'Dashwood',\n",
       " 'had',\n",
       " 'long',\n",
       " 'been',\n",
       " 'settled',\n",
       " 'in',\n",
       " 'Sussex',\n",
       " '.']"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sent1 + sent2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "关于NLTK中文本数据集的操作先介绍到这，文本相关的操作即操作字符串，这和Python操作字符串是相同的。"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 关于简单的统计"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "我们已经了解了Python中的列表和字符串的相关操作，我们即可以进行一些简单的语言分析了。"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 频率分布"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "我们想了解一个文本，想知道它是什么风格的文章，我们需要对文章用词进行分析，设想一下，如何知道一个文本里使用最频繁的50个字是什么？我们可以考虑写一个计数器，或者，使用NLTK内置的函数。"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "接下来我们来找出《白鲸记》中最常见的50个词。NLTK内置的text1文本即《白鲸记》。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "fdist1 = FreqDist(text1)   # 使用FreqDist统计每个字符出现的次数，key是字符，value为出现的次数"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "FreqDist({',': 18713, 'the': 13721, '.': 6862, 'of': 6536, 'and': 6024, 'a': 4569, 'to': 4542, ';': 4072, 'in': 3916, 'that': 2982, ...})"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fdist1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 书中使用的是这种方式，但是错误明显，不知道是不是nltk更新版本的原因\n",
    "# fdist1.keys()[:50]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(',', 18713),\n",
       " ('the', 13721),\n",
       " ('.', 6862),\n",
       " ('of', 6536),\n",
       " ('and', 6024),\n",
       " ('a', 4569),\n",
       " ('to', 4542),\n",
       " (';', 4072),\n",
       " ('in', 3916),\n",
       " ('that', 2982),\n",
       " (\"'\", 2684),\n",
       " ('-', 2552),\n",
       " ('his', 2459),\n",
       " ('it', 2209),\n",
       " ('I', 2124),\n",
       " ('s', 1739),\n",
       " ('is', 1695),\n",
       " ('he', 1661),\n",
       " ('with', 1659),\n",
       " ('was', 1632),\n",
       " ('as', 1620),\n",
       " ('\"', 1478),\n",
       " ('all', 1462),\n",
       " ('for', 1414),\n",
       " ('this', 1280),\n",
       " ('!', 1269),\n",
       " ('at', 1231),\n",
       " ('by', 1137),\n",
       " ('but', 1113),\n",
       " ('not', 1103),\n",
       " ('--', 1070),\n",
       " ('him', 1058),\n",
       " ('from', 1052),\n",
       " ('be', 1030),\n",
       " ('on', 1005),\n",
       " ('so', 918),\n",
       " ('whale', 906),\n",
       " ('one', 889),\n",
       " ('you', 841),\n",
       " ('had', 767),\n",
       " ('have', 760),\n",
       " ('there', 715),\n",
       " ('But', 705),\n",
       " ('or', 697),\n",
       " ('were', 680),\n",
       " ('now', 646),\n",
       " ('which', 640),\n",
       " ('?', 637),\n",
       " ('me', 627),\n",
       " ('like', 624)]"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 我改写了一下，先将FreqDist排序输出为列表，再取前50个值\n",
    "n = sorted(fdist1.items(), key=lambda x:x[1], reverse=True)\n",
    "n[:50]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "13721"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fdist1['the']    # 查看某个词使用的次数"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "我们可以通过图表的方式来看一下这前50个词占据了整本书的多比例"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x13c9a4d10>"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fdist1.plot(50, cumulative=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "260819"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(text1)    # 整本书的词量"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "可以看到这些词占了书本将近一半的词量，这样的话，这些高频词对于我们的分析没有多大帮助，数量多，且都是管道词语，没有信息量。\n",
    "\n",
    "既然高频词没有研究价值，那低频词呢？我们来看看那些在文本中只出现了一次的词。我们使用NLTK内置的`hapaxes`来查看。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Herman',\n",
       " 'Melville',\n",
       " ']',\n",
       " 'ETYMOLOGY',\n",
       " 'Late',\n",
       " 'Consumptive',\n",
       " 'School',\n",
       " 'threadbare',\n",
       " 'lexicons',\n",
       " 'mockingly']"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "low = fdist1.hapaxes()    # 将所有低频词放入列表low中\n",
    "low[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fdist1['Herman']    # 验证一下，Herman这个词确实只有一个"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "9002"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(low)    # 这样的低频词有9002个"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "低频词的数量依然很多。"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 细粒度的词分析"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "既然高频词和低频词都无法进行很好的分析，接下来要想其他办法了；\n",
    "我们来看看书中的`长词`，例如那些长度超过15的词。"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "我们定义一个性质P，P(w)为真当且仅当词w的长度大于15。我们用数学集合来表示："
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$$ \\lbrace w \\quad | \\quad w \\quad \\in \\quad V \\quad \\& \\quad P(w) \\rbrace $$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "使用Python怎么表示呢？"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$$ [w \\quad for \\quad w \\quad in \\quad V \\quad if \\quad P(w)] $$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "举个例子："
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "V = set(text1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "long_words = [w for w in V if len(w) > 15]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['CIRCUMNAVIGATION',\n",
       " 'Physiognomically',\n",
       " 'apprehensiveness',\n",
       " 'cannibalistically',\n",
       " 'characteristically',\n",
       " 'circumnavigating',\n",
       " 'circumnavigation',\n",
       " 'circumnavigations',\n",
       " 'comprehensiveness',\n",
       " 'hermaphroditical',\n",
       " 'indiscriminately',\n",
       " 'indispensableness',\n",
       " 'irresistibleness',\n",
       " 'physiognomically',\n",
       " 'preternaturalness',\n",
       " 'responsibilities',\n",
       " 'simultaneousness',\n",
       " 'subterraneousness',\n",
       " 'supernaturalness',\n",
       " 'superstitiousness',\n",
       " 'uncomfortableness',\n",
       " 'uncompromisedness',\n",
       " 'undiscriminating',\n",
       " 'uninterpenetratingly']"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sorted(long_words)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 词语搭配和双连词"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "词语间存在固定的搭配，就像`红酒`，而如果是`棕酒`，就很奇怪；我们使用`bigrams（）`函数查看双连词。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'red': 'wine', 'wine': 'black', 'black': 'board'}"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "words = ['red', 'wine', 'black', 'board']\n",
    "dict(bigrams(words))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "ename": "ValueError",
     "evalue": "too many values to unpack (expected 2)",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-40-61a61879b276>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0;31m# 原文使用collocations，报错\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m \u001b[0mtext8\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcollocations\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m/opt/anaconda3/envs/blog/lib/python3.7/site-packages/nltk/text.py\u001b[0m in \u001b[0;36mcollocations\u001b[0;34m(self, num, window_size)\u001b[0m\n\u001b[1;32m    442\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    443\u001b[0m         collocation_strings = [\n\u001b[0;32m--> 444\u001b[0;31m             \u001b[0mw1\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0;34m\" \"\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mw2\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mw1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mw2\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcollocation_list\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnum\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mwindow_size\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    445\u001b[0m         ]\n\u001b[1;32m    446\u001b[0m         \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtokenwrap\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcollocation_strings\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mseparator\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m\"; \"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/anaconda3/envs/blog/lib/python3.7/site-packages/nltk/text.py\u001b[0m in \u001b[0;36m<listcomp>\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m    442\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    443\u001b[0m         collocation_strings = [\n\u001b[0;32m--> 444\u001b[0;31m             \u001b[0mw1\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0;34m\" \"\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mw2\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mw1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mw2\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcollocation_list\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnum\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mwindow_size\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    445\u001b[0m         ]\n\u001b[1;32m    446\u001b[0m         \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtokenwrap\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcollocation_strings\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mseparator\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m\"; \"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mValueError\u001b[0m: too many values to unpack (expected 2)"
     ]
    }
   ],
   "source": [
    "# 原文使用collocations，报错\n",
    "text8.collocations()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'weekday': 'weekend',\n",
       " 'weekend': 'weekends',\n",
       " 'weekends': 'welcome',\n",
       " 'welcome': 'well',\n",
       " 'well': 'western',\n",
       " 'western': 'when',\n",
       " 'when': 'who',\n",
       " 'who': 'widowed',\n",
       " 'widowed': 'wild',\n",
       " 'wild': 'wine',\n",
       " 'wine': 'wining',\n",
       " 'wining': 'with',\n",
       " 'with': 'without',\n",
       " 'without': 'woman',\n",
       " 'woman': 'women',\n",
       " 'women': 'work',\n",
       " 'work': 'working',\n",
       " 'working': 'would',\n",
       " 'would': 'write',\n",
       " 'write': 'y',\n",
       " 'y': 'year',\n",
       " 'year': 'years',\n",
       " 'years': 'yo',\n",
       " 'yo': 'you',\n",
       " 'you': 'young',\n",
       " 'young': 'younger',\n",
       " 'younger': 'your',\n",
       " 'your': 'yr',\n",
       " 'yr': 'yrs'}"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 改用另一种方式\n",
    "dict(bigrams(sorted(set(list(text8)))[-30:]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 其他类型的计数统计"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "通过Python的列表可统计文本中词的长度"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[2,\n",
       " 4,\n",
       " 4,\n",
       " 1,\n",
       " 5,\n",
       " 6,\n",
       " 5,\n",
       " 6,\n",
       " 4,\n",
       " 1,\n",
       " 3,\n",
       " 8,\n",
       " 10,\n",
       " 1,\n",
       " 4,\n",
       " 8,\n",
       " 5,\n",
       " 1,\n",
       " 7,\n",
       " 4,\n",
       " 2,\n",
       " 7,\n",
       " 3,\n",
       " 3,\n",
       " 5,\n",
       " 1,\n",
       " 2,\n",
       " 2,\n",
       " 6,\n",
       " 3]"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 这里缩短到了前30个\n",
    "[len(w) for w in text8][:30]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "FreqDist({1: 1455, 4: 752, 3: 565, 2: 555, 5: 458, 6: 373, 7: 306, 8: 148, 10: 88, 9: 79, ...})"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 词长度统计\n",
    "fdist1 = FreqDist([len(w) for w in text8])\n",
    "fdist1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys([2, 4, 1, 5, 6, 3, 8, 10, 7, 12, 11, 9, 13, 14])"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 一共有哪些长度的词\n",
    "fdist1.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_items([(2, 555), (4, 752), (1, 1455), (5, 458), (6, 373), (3, 565), (8, 148), (10, 88), (7, 306), (12, 44), (11, 37), (9, 79), (13, 6), (14, 1)])"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fdist1.items()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 哪个词长数量最多\n",
    "fdist1.max()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.2989521265666735"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 长度1的词最多，占了全书29.9%\n",
    "fdist1.freq(1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 更多常用方法"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "方法|介绍\n",
    "- | -\n",
    "fdist = FreqDist(samples)|创建样本的频率分布|\n",
    "fdist.inc(sample)|增加样本|\n",
    "fdist.freq('monstrous')|计数词出现的次数\n",
    "fdist.N()|样本词总数\n",
    "fdist.keys()|频率数递减排序\n",
    "fdist.max()|数值最大的词（样本）\n",
    "fdist.tabulate()|频率分布表\n",
    "fdist.plot()|频率分布图"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   1    4    3    2    5    6    7    8   10    9   12   11   13   14 \n",
      "1455  752  565  555  458  373  306  148   88   79   44   37    6    1 \n"
     ]
    }
   ],
   "source": [
    "fdist1.tabulate()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x13c0d9a10>"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fdist1.plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 如果您看到了这里，请给一个star，谢谢，如果有遗漏，请留言指正，感激！"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "blog",
   "language": "python",
   "name": "blog"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}