Mendenhall-CharacteristicCurve.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Mendenhall's Characteristic Curve (1887): Early Stylometrics"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In 1887 the polymath T. C. Mendenhall published an article in Science titled, \"The Characteristic Curves of Composition\" which is both one of the earliest examples of quantitative stylistics but also one of the first studies to present text visualizations based on the (manual) count of words. Mendenhall thought that different authors would have distinctive curves of word length frequencies which could help with authorship attribution. \n",
    "\n",
    "Here you can see an example of the characteristic curve of _Oliver Twist_. Mendenhall took the first 1000 words, counted the length in characters of these 1000 words and then graphed the number of words of each length. Thus one can see that there is just under 50 words of one letter length in the first one thousand words.\n",
    "\n",
    "![Mendhall Characteristic Curve](data/Mendenhall-CharacteristicCurve/OliverTwist-CharacteristicCurve.png)\n",
    "\n",
    "Mendenhall thought this method of analysis would help with the \"identification or discrimination of authorship\" or authorship attribution as we call it today."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Acquiring the Text"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We'll begin by fetching the edition of [Oliver Twist](http://www.gutenberg.org/cache/epub/730/pg730.txt) that's available from the [Gutenberg Project](http://en.wikipedia.org/wiki/Project_Gutenberg). It's always worth noting that other editions are available and may be preferable for various reasons, but the Gutenberg edition has the benefit of convenience and longevity.\n",
    "\n",
    "Fetching a URL can be very quick and easy in Python, but we'll take a detour to first check if there's a local copy of the text file – this may help avoid some problems of connectivity and access management."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os, urllib.request\n",
    "\n",
    "# get the text from a filename (if it exists) or fetch the contents (and cache it for next time)\n",
    "def get_text_local_or_url(filename, url):\n",
    "    \n",
    "    # do we have a local copy? if so read it in and return it\n",
    "    if os.path.exists(filename):\n",
    "        with open(filename) as f:\n",
    "            return f.read()\n",
    "    \n",
    "    # if not, fetch the URL contents and then write it out for next time\n",
    "    else:\n",
    "        string = urllib.request.urlopen(url).read().decode()\n",
    "        with open(filename, \"w\") as f:\n",
    "            f.write(string)\n",
    "        return string"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So now we're ready to make a call to fetch the text."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "﻿The Project Gutenberg EBook of Oliver Twist, by C [...]  to our email newsletter to hear about new eBooks.\n"
     ]
    }
   ],
   "source": [
    "oliverTwistAllString = get_text_local_or_url(\"data/Mendenhall-CharacteristicCurve/OliverTwist.txt\",\n",
    "    \"http://www.gutenberg.org/cache/epub/730/pg730.txt\").strip()\n",
    "\n",
    "# show a snippet\n",
    "print(oliverTwistAllString[:50], \"[...]\", oliverTwistAllString[-50:])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The Gutenberg text includes some header and footer text, including licensing information, that we want to strip out before doing our analysis. In looking at the [original text](http://www.gutenberg.org/cache/epub/730/pg730.txt), we can see that see that things start with \"CHAPTER I\" (assuming we don't want the table of contents). Likewise, the end of the text, before the file footer, can be found with \"End of the Project Gutenberg EBook of Oliver Twist\". So we find the locations of those strings and take what's in between."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CHAPTER I\n",
      "\n",
      "TREATS OF THE PLACE WHERE OLIVER TWIST  [...]  nook is in a Church, and she was weak and erring.\n"
     ]
    }
   ],
   "source": [
    "start = oliverTwistAllString.find(\"CHAPTER I\")\n",
    "end = oliverTwistAllString.find(\"End of the Project Gutenberg EBook of Oliver Twist\")\n",
    "oliverTwistString = oliverTwistAllString[start:end].strip()\n",
    "\n",
    "# show a snippet\n",
    "print(oliverTwistString[:50], \"[...]\", oliverTwistString[-50:])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We have the text in good shape now, the next step is to analyze it."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Analyzing Word Lengths"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's talk through what we want to do with our string to analyze the word length:\n",
    "\n",
    "* break the one text string into a list of words\n",
    "* measure the length of each word\n",
    "* graph the word lengths (in order of length, not frequencies)\n",
    "\n",
    "We can start with the first task, breaking the string into words, or tokenization."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['CHAPTER', 'I', 'TREATS', 'OF', 'THE']"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import nltk\n",
    "\n",
    "tokens = nltk.word_tokenize(oliverTwistString)\n",
    "words = [token for token in tokens if any(c for c in token if c.isalpha())]\n",
    "\n",
    "# preview first five words\n",
    "words[:5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next we want to measure the length of each word. Note that Mendenhall was counting words manually and he only considered the first 1,000 words, so we'll take a subset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('CHAPTER', 7), ('I', 1), ('TREATS', 6), ('OF', 2), ('THE', 3)]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lengths = [len(word) for word in words[:1000]]\n",
    "\n",
    "# preview the first ten aligned words and lengths\n",
    "list(zip(words,lengths))[:5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Then we want to count the word lengths."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(1, 38), (2, 166), (3, 231), (4, 179), (5, 125)]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lengthFreqs = nltk.FreqDist(lengths)\n",
    "orderedLengthFreqs = sorted(list(lengthFreqs.items()))\n",
    "\n",
    "# preview the first 5 values\n",
    "orderedLengthFreqs[:5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally, we'll plot the curve."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEKCAYAAAAIO8L1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvOIA7rQAAIABJREFUeJzt3Xl8FeXVwPHfyb4QlpCAbCHsmwqEoKxaF6xaKy7VYutWRbTVLmoXtb5trX1bq6+1Wq0VlxZprWtt3RfUyq4kIIvsZIEAJoGEJCSQ9bx/3AmNNoQbuHPnLuf7+czn3pnMnefcwM258zwz5xFVxRhjjPmiGK8DMMYYE5osQRhjjGmXJQhjjDHtsgRhjDGmXZYgjDHGtMsShDHGmHZZgjDGGNMuSxDGGGPaZQnCGGNMu+K8DuBYZGRkaHZ2ttdhGGNMWMnPz9+jqplH2i+sE0R2djZ5eXleh2GMMWFFRIr92c+6mIwxxrTLEoQxxph2WYIwxhjTLksQxhhj2mUJwhhjTLssQRhjjGmXJQhjjDHtsgQR5jZ9VsO760u9DsMYE4EsQYQxVeXWFz7h+vl5bC6t8TocY0yEsQQRxpYV7GXdzmpaFH71+gavwzHGRBhLEGHs8YUFZHRJ4EdfHsHCzeV8sKnM65CMMRHEEkSY2lxawwebyrlycjbXTR/MoIxUfvXaehqbW7wOzRgTISxBhKknFhWQFB/D5ZMGkhAXwx3njmJbeS3PfLTd69CMMRHCEkQYKqs+yD9X7eKSCQNIT00A4MxRvZgypCcPLNhMVV2jxxEaYyKBJYgw9JelRTS2tDB7+qBD20SEO78ymqoDjTz0/hYPozPGRApLEGGmtr6Jvy4v5uwxxzGwZ+rnfja6b1dmTRzAvKVFFJTv9yhCY0yksAQRZp7P20H1wSauO2Vwuz+/ZcYIkuJj+fUbG4McmTEm0liCCCNNzS08ubiQ3IE9yMnq0e4+mWmJfOe0ISzYUMqSrXuCHKExJpJYgggjb677jJLKA8w5zNlDq2umDqJ/j2Tufm09zS0apOiMMZHGEkSYUFXmLixgUEYqZ47q3eG+SfGx3H7OKDZ+VsPzeTuCFKExJtJYgggTHxVWsHZnFbOnDyImRo64/7knHMfE7B7c/84mag7aZa/GmM6zBBEmHl9YQHpqAhfn9Pdr/9bLXvfsb+CP/97mcnTGmEhkCSIMbCmt4b2NZVw5eSBJ8bF+v27sgO5clNOPJxcVsqOizsUIjTGRyBJEGHhiUSGJcTFcOTm706/98ZdHEhsj3POmXfZqjOkcSxAhrqzmIC+v2skluf0PldXojOO6JXH9qYN5fe1uVhRVuBChMSZSWYIIcU8vLaaxpYVrp3V8aWtHrj9lCH26JfHLV9fTYpe9GmP8ZAkihNXWNzF/eTFnje7NoIzUI7/gMJITYvnx2SNYu7OKl1ftDGCExphIZgkihL2Qt4OqA43MOWXIMR9r5th+jO3fjXvf3khdQ1MAojPGRDpLECGqqbmFJ5cUMmFgDyYMbL+sRmfExAg/++poSqvr+dOHBQGI0BgT6SxBhKi3Py1lR8UBrpt+9GMPXzRhYDrnndiHuQu3sbvqQMCOa4yJTJYgQpCvrMY2snumMGN0x2U1Ouu2c0bSonDvW5sCelxjTORxLUGIyAAR+UBENojIpyLyfWd7uoi8KyJbnMceznYRkYdEZKuIrBGRHLdiC3UfF1awuqSK2dMHE+tHWY3O6N8jhdnTBvHyqp18smNfQI9tjIksbp5BNAG3quooYBJwo4iMBm4D3lPVYcB7zjrAOcAwZ5kDPOpibCHt8UWdK6vRWd85bSgZXRK5+7X1qNplr8aY9rmWIFR1t6qudJ7XABuAfsBMYJ6z2zzgAuf5TOBp9VkOdBeRPm7FF6q2lu1nwYYyrpg0kOQE/8tqdEaXxDh+9OXh5BdX8tqa3a60YYwJf0EZgxCRbGA88BHQW1V3gy+JAL2c3foBbWtTlzjbosqTiwtIjIvhiskDXW3naxMGMLpPV+55cyMHG5tdbcsYE55cTxAi0gV4CfiBqlZ3tGs72/6r/0NE5ohInojklZeXByrMkFBeU89LK3dy8YT+ZHRJdLWt2BjhzvNGsXPfAZ5cXOhqW8aY8ORqghCReHzJ4W+q+g9nc2lr15HzWOZsLwEGtHl5f2DXF4+pqnNVNVdVczMzM90L3gNPLyuisbmF2dMGBaW9KUMyOGt0b/74wVbKag4GpU1jTPhw8yomAZ4ENqjq79r86BXgKuf5VcC/2my/0rmaaRJQ1doVFQ3qGnxlNWaM6s3gzC5Ba/eOc0fR0NzC/W9vDlqbxpjw4OYZxFTgCuB0EfnEWc4F7gFmiMgWYIazDvAGUABsBR4HvuNibCHnxfwS9tU1HnG+6UDLzkjl6inZPJ+/g3U7q4LatjEmtMW5dWBVXUz74woAZ7SzvwI3uhVPKGtuUZ5YVMj4rO4BKavRWTedPoyXVu7kV6+v5+/XTcJ38meMiXZ2J3UIePvTz9heUcf1pwz25I9zt+R4bj5zGMsLKnhnfWnQ2zfGhCZLEB5TVR5bWMDAninMGH2cZ3FcdlIWw3p14ddvbKC+yS57NcZYgvBcXnElq3fsY/a0QQEvq9EZcbEx3HneaIr31vH00mLP4jDGhA5LEB577MMCeqTE87UJA468s8tOHZ7Jl0Zk8tD7W9i7v97rcIwxHrME4aFt5ftZsKGUKyZnu1ZWo7Pu/Moo6hqa+f2CLV6HYozxmCUIDz2xqJCEuBiudLmsRmcM7ZXG5Sdn8bePitlcWuN1OMYYD1mC8Mie/fW8tLKEi3PcL6vRWT84czhdEuP41esbvA7FGOMhSxAeeXpZMQ1NLcyeHpyyGp3RIzWB750xjIWby/lgU9mRX2CMiUiWIDxwoKGZ+cuKOHNUb4YEsaxGZ1w5OZtBGan87+sbaGxu8TocY4wHLEF44MX8HVTWNXL9qcEtq9EZCXEx3HHuKLaW7eeZj7Z7HY4xxgOWIIKsuUV5YnEh4wZ0J9eDshqdceaoXkwZ0pMHFmymqq7R63CMMUFmCSLI3l3/GcV765jjUVmNzhAR7vzKaKoONPLQ+3bZqzHRxhJEELWW1chKT+HLY7wrq9EZo/t2ZdbEAcxbWkRB+X6vwzHGBJEliCDKL65k1fZ9zJ7ubVmNzrplxggS42K47+1NXodijAkiSxBBNHdhAd1T4vnahP5eh9IpmWmJzDllCG+u+4z84kqvwzHGBIkliCApKN/PuxtKuWLSQFISXJuGwzWzpw8iMy2Re97cgG/qDmNMpLMEESRPLi4kPiaGKydnex3KUUlNjOPmM4ezoqiSd23OCGOigiWIINizv54X80u4KKcfmWmhVVajMy7N7c+QzFTueWsjTXbznDERzxJEEMxfVkx9Uwuzp4fujXH+iIuN4bZzRlFQXstzeTu8DscY4zJLEC470NDM/OXFnDmqF0N7hWZZjc44c1QvTspO54F3t1Bb3+R1OMYYF1mCcNlLK0uoqG3gujA/e2glItx+7kj27K/n8UUFXodjjHGRJQgXNbcoTywqYGz/bpw0KN3rcAJmfFYPvnJCH+YuLKCs5qDX4RhjXGIJwkXvri+laG8dc04ZEvJlNTrrR18eQUNTCw/azHPGRCxLEC56fFEBA9KT+fKY3l6HEnDZGal88+Qsnl2xg61lVoLDmEhkCcIle/bXk19cyayJWcTFRuav+btnDCM5PpZ739rodSjGGBdE5l+uENBakuLkCBp7+KKMLonccOpg3llfyoqiCq/DMcYEmCUIl+QXV5IQG8Px/bp5HYqrrp02mN5dE/n1G1aCw5hIYwnCJXlFFZzQvxtJ8bFeh+Kq5IRYbpkxnFXb9/HWus+8DscYE0CWIFxwsLGZdTurQ37GuEC5OKc/w3t34d63N9n81cZEEEsQLli7s4qG5hYmREmC8JXgGEnhnlqe/djmrzYmUliCcEFekW+AOloSBMBpI3oxaXA6v1+whf1WgsOYiGAJwgX5xRUMzkilZ5fwrdzaWSLC7eeMYm9tA3M/3OZ1OMaYALAEEWCqSn5xZVSdPbQaO6A7Xx3bl8cXFVJabSU4jAl3liACbFt5LZV1jeRmR1+CAPjRWSNoamnh9ws2ex2KMeYYuZYgROQpESkTkXVttv1CRHaKyCfOcm6bn90uIltFZJOIfNmtuNyWX+y7YWzCwMi9Qa4jWT1TuHzSQJ5bsYMtpTVeh2OMOQZunkH8BTi7ne0PqOo4Z3kDQERGA7OAMc5r/igiYXkDQV5RJT1S4hmSmep1KJ757unDSE2I47dWgsOYsOZaglDVhYC/9RdmAs+qar2qFgJbgZPcis1NreMPkVa9tTPSUxP49mlDWLChjOUFe70OxxhzlLwYg7hJRNY4XVCtHfX9gLZzWJY428LK3v31FOypjdrupbaumTqIPt2S+I2V4DAmbAU7QTwKDAHGAbuB+53t7X3dbveviojMEZE8EckrLy93J8qj1FqgL1oHqNtKiveV4FhdUsXra3d7HY4x5igcMUGIyHx/tvlDVUtVtVlVW4DH+U83UgkwoM2u/YFdhznGXFXNVdXczMzMownDNa0F+k6I8AJ9/roopz8jj0vj3rc20dBkJTiMCTf+nEGMabviDB5POJrGRKRPm9ULgdYrnF4BZolIoogMAoYBHx9NG15aUVTB8f26RnyBPn/Fxgi3nTOS7RV1PPNRsdfhGGM66bAJwrnstAY4UUSqnaUGKAP+daQDi8jfgWXACBEpEZFrgXtFZK2IrAFOA24GUNVPgeeB9cBbwI2q2nysby6YDhXoy7bxh7ZOHZ7J1KE9eej9rVQfbPQ6HGNMJxw2Qajqb1Q1DbhPVbs6S5qq9lTV2490YFW9TFX7qGq8qvZX1SdV9QpVPUFVT1TV81V1d5v9/1dVh6jqCFV9M0DvL2iirUCfv1pLcFTUNvCYleAwJqwcsYtJVW8XkX4iMkVETmldghFcOInGAn3+Or5fNy4Y15cnFhWyu+qA1+EYY/zkzyD1PcAS4E7gR87yQ5fjCjv5xRUMykglI4oK9HXGrWeNQBUeeNdKcBgTLuL82OdCYISq1rsdTLhqLdB3xqjeXocSsgakp3Dl5IE8taSQa6YNYuRxXb0OyRhzBP5cxVQAxLsdSDhrLdA30e5/6NBNpw+lS2Icv33TSnAYEw78OYOoAz4RkfeAQ2cRqvo916IKM9FeoM9f3VMSuPG0ofzmzY0s3bqHKUMzvA7JGNMBf84gXgHuBpYC+W0W47ACff67ako2/bon85s3N9LSYiU4jAllRzyDUNV5wQgknFmBPv8lxcdy61nDueX51by6Zhczx4VdyS1jooY/VzEVikjBF5dgBBcOrEBf510wrh+j+nTlvrc3Ud8UVvdDGhNV/OliygUmOst04CHgr24GFU6sQF/nxcQId5w7kpLKA8xfZiU4jAlV/twot7fNslNVfw+cHoTYwoIV6Ds604dlMn1YBg9/sJWqA1aCw5hQ5E8XU06bJVdEbgDSghBbWMgrrrQCfUfptnNGUnWgkUf/bSU4jAlF/lzmen+b501AEXCpK9GEmYONzawtqeLqqdlehxKWxvTtxoXj+/HUkkKumDyQft2TvQ7JGNOGP11Mp7VZZqjqdaq6KRjBhbp1VqDvmN161ggAfveOleAwJtT408XUTUR+1zqLm4jcLyLW4Y6vewmsQN+x6Nc9mW9NyeYfq0pYv6va63CMMW34cxXTU0ANvm6lS4Fq4M9uBhUu8ooqrUBfAHznS0PpmhTPPW9ZCQ5jQok/CWKIqv5cVQuc5S5gsNuBhTpVZeX2Sjt7CIBuKfF89/ShLNxczqItoTXPuDHRzJ8EcUBEprWuiMhUIOqL+hfsqaWitoFcSxABccXkgWSlp/CjF9ZQWn3Q63CMMfiXIG4AHhGRIhEpAh52tkW1/CK7QS6QEuNi+dPlE6g+2Mh1T+dxoMHusDbGa/5cxbRaVccCJwInqup4VV3tfmihLa+4gu4p8QzO6OJ1KBFjdN+uPDhrPGt3VnHrC59YMT9jPHbYBCEit4jIta3rqlqtqtUi8l0R+UFwwgtdecWVTMjqQUyMFegLpBmje3PHOaN4Y+1n/M5mnzPGUx2dQVwDzG9n+1znZ1GroraBgvJaJlj3kitmTx/ErIkDePiDrby8qsTrcIyJWh0lCFXVhnY21gNR/bX5UIE+q+DqChHhlzOPZ/LgnvzkxbXkFVV4HZIxUanDMQgR+a9JltvbFm3yiiuIjxVO7G/3C7olIS6GRy/PoV+PZObMz2f73jqvQzIm6nSUIO4DXheRU0UkzVm+BLwK/F9QogtR+UWVHN+vmxXoc1n3lASevCqX5hbl2nkrqD5oVV+NCabDJghVfRr4H+CX+Ar0FQJ3AT+P5lnm6puaWbOzyu5/CJLBmV149PIcCvfUctMzq2hqbvE6JGOiRoddTKr6pqqeqqo9VTXDef5msIILRet2VtHQ1GIzyAXRlCEZ/OqC41m4uZy7X1vvdTjGRA1/yn2bNvKKrECfF2adlMW28v08vqiQIb26cOXkbK9DMibiWYLopLziSrJ7ppCZZgX6gu22c0ZRuKeWu15dz8CeqZw6PNPrkIyJaP6U2jAOVWVlcaV1L3kkNkZ4cNZ4hvdO46a/rWRLaY3XIRkT0Q57BiEit3T0QlX9XeDDCW2Fe2rZW9tg9Zc8lJoYx5NX5TLzkSVcM28F//zOVHpauXVjXNHRGUSas+QC3wb6OcsNwGj3Qws9eYdukLME4aW+3ZN5/MpcyqrruX5+PvVNVtjPGDd0dJnrXc7cDxlAjqreqqq3AhOA/sEKMJTkF1XSLTmeIZlWoM9r4wZ05/5Lx5JXXMntL61F1Qr7GRNo/gxSZwFtS240ANmuRBPi8oormDDQCvSFivNO7EtBeS2/e3czQ3p14cbThnodkjERxZ8EMR/4WEReBhS4EIi6G+UqaxvYVl7LRTlRefIUsr57+lC2le/nvrc3MTgjlXNO6ON1SMZEDH/mg/hf4FtAJbAP+Jaq/uZIrxORp0SkTETWtdmWLiLvisgW57GHs11E5CER2Soia0Qk5+jfkjvybfwhJIkIv734RHKyunPz85+wpmSf1yEZEzGOVKwvRkTWqepKVX3QWVb5eey/AGd/YdttwHuqOgx4z1kHOAcY5ixzgEf9fQPBkldcSXysMHZAd69DMV+QFB/L3CtzyeiSyOx5eeyuivoZcY0JiCOV2mgBVotIVmcPrKoLgS/WaZ7Jf7qn5gEXtNn+tPosB7qLSEj1FeQXVzCmrxXoC1UZXRJ58qqJ1DU0M3teHnUNTV6HZEzY8+dGuT7ApyLynoi80rocZXu9VXU3gPPYy9neD9jRZr8SZ1tIqG9qZnWJFegLdSOOS+MPl41nw+5qfvCsTVlqzLHyZ5D6LtejaH8ConY/3SIyB183FFlZnT6xOSrrdlbT0NRiN8iFgdNG9uJ/zhvNXa+u5963N3HbOSO9DsmYsOXPIPWHwEb+c+PcBmfb0Sht7TpyHsuc7SXAgDb79Qd2HSaeuaqaq6q5mZnBqcWTX+zrKbMSG+Hh6inZXD4piz99uI3n83Yc+QXGmHYdMUGIyKXAx8AlwKXARyLytaNs7xXgKuf5VcC/2my/0rmaaRJQ1doVFQryiioZaAX6woaI8POvjmH6sAx++vJalhfs9TokY8KSP2MQPwUmqupVqnolcBK+iYQ6JCJ/B5YBI0SkRESuBe4BZojIFmCGsw7wBlAAbAUeB77T6XfiElUlv7jSynuHmfjYGB7+Rg5Z6Snc8Nd8ivbUeh2SMWHHnwQRo6plbdb3+vM6Vb1MVfuoaryq9lfVJ1V1r6qeoarDnMcKZ19V1RtVdYiqnqCqeUf5fgKuaG+dr0CfdS+FnW7J8Tx19UQEuGbeCqrqbMpSYzrDnwTxloi8LSJXi8jVwOv4vvFHhbwi3/iDDVCHp4E9U3nsilx2VNRx4zMrabQpS43xmz9nAj8C5gInAmOBuar6E7cDCxX5xZV0TYpjqBXoC1snDUrn1xeewOKte/j5K59aYT9j/NTRfBA/AJYAq1T1JeCloEUVQvKKK8nNTrcCfWHuktwBFOyp5dF/b2NIZheunTbI65CMCXkd3QfRH3gQGCkia4Cl+BLGstaxg0hXWdvA1rL9XDg+ZO7ZM8fgR2eNoLC8lrtfW09CrHCFzWttTIc6mg/ih6o6BTgOuANf2YxrgHUisj5I8XnKCvRFlpgY4fezxnHmqF78z78+5U8fbvM6JGNCmj+D1MlAV6Cbs+wCPnIzqFBhBfoiT1J8LI9ePoGvju3LPW9u5P53NtmYhDGH0dEYxFxgDFCDLyEsBX6nqpVBis1zVqAvMsXHxvD7r48jNSGWP7y/lf31TfzsvNGI2DiTMW11NAaRBSQCW4Cd+MphRE2x/dYCfVdOGuh1KMYFsTHCby46gZSEOJ5aUkhdfTO/vugEYu1iBGMOOWyCUNWzxfeVagwwBbgVOF5EKvANVP88SDF6wgr0RT4R4X/OG0WXxFgeen8rtQ1NPPD1ccTH+tPzakzk67Caq/o6Z9eJyD6gylnOw1duI6IThBXoiw4iwi1njSA1MY7fvLmRAw3NPPLNHOtWNIYOBqlF5Hsi8qyI7AAW4ksMm4CLgIj/q2kF+qLL9acO4VcXHM/7m8q45i8rqK23CYeM6egMIht4Ebg5lCqrBkNrgb5TRwSnnLgJDZdPGkhqYiw/fGENVzz5EX/+1kl0S473OixjPNPRfRC3qOqL0ZYcwAr0RbMLx/fnkW/ksHZnFZfNXc7e/fVeh2SMZ2w0rh1WoC+6nX38cTxx1UQK9uzn0seW8VnVQa9DMsYTliDaYQX6zKnDM5n3rZMora7nkseWsn1vndchGRN0liDakedMEGQF+qLbyYN78rfZJ1NzsIlLHlvK1rIar0MyJqgsQXzBvjpfgb7cbBt/MDB2QHeemzOZ5ha49LHlrNtZ5XVIxgSNJYgvaC3QZ1OMmlYjjkvjhRsmkxQXw2WPLz/0f8SYSGcJ4gvyiiuJixHG9rcCfeY/BmWk8sK3p9AzNYErnvyIJVv3eB2SMa6zBPEF+UWVjOnXjeQEu5PWfF6/7sk8f8NkBvRI4Vt/WcGC9aVeh2SMqyxBtNHQ1MLqkn02/4M5rF5pSTx3/SRGHZfGDX/N55XVu7wOyRjXWIJoY92uKuqbWixBmA51T0ngr7NPJmdgD77/7CqeW7Hd65CMcYUliDbyi5wBartBzhxBWlI88751EqcMy+QnL63lqcWFXodkTMBZgmgjr7iCrPQUeqUleR2KCQPJCbHMvXICZ485jl++tp4/vLfFZqczEcUShKO1QJ91L5nOSIyL5eFvjOei8f24/93N3PPWRksSJmJ0OB9ENCneW8ee/Q3WvWQ6LS42hv+7ZCwpibE89mEBtfVN/PL84+1OfBP2LEE48pybn6yCqzkaMTHC3TOPJzUxjsc+LKCuvpl7v3YicTY7nQljliAc+cUVdE2KY1gvK9Bnjo6IcNvZI0lLjOP/3tlMac1BHr4shx6pCV6HZsxRsa83jryiSnKsQJ85RiLCTacP496vnciKwkpmPrKEzaVW5M+EJ0sQ+Ar0bSnbbwPUJmAuzR3As9dP4kBjMxc+soS3P/3M65CM6TRLEMDK7a0F+mz8wQROTlYPXr1pGkN7p3H9/HweXLCFlha7wsmED0sQ+LqX4mKEcQOsQJ8JrOO6JfHcnElclNOPBxZs5jt/W0ltfZPXYRnjF0sQ+K5gGtO3qxXoM65Iio/l/kvGcudXRvHO+s+4+NGl7KiwGepM6Iv6BNHQ1MLqHfuse8m4SkSYPX0w8645id1VBzn/4cUstZLhJsR5kiBEpEhE1orIJyKS52xLF5F3RWSL8xiUEeNPWwv02Q1yJgimD8vkXzdOJaNLIlc89TF/WVJod16bkOXlGcRpqjpOVXOd9duA91R1GPCes+66/EM3yFmCMMGRnZHKyzdO5fSRvfjFq+v5yUtrqG9q9josY/5LKHUxzQTmOc/nARcEo9G8okoGpCfTq6sV6DPB0yUxjscun8D3Th/K83klXDZ3OWXVB70Oy5jP8SpBKPCOiOSLyBxnW29V3Q3gPPZyPQhV8oorrbyG8URMjHDLWSP44zdz2LC7hvMfXsLqHfu8DsuYQ7xKEFNVNQc4B7hRRE7x94UiMkdE8kQkr7y8/JiC2F5Rx5799Uyw7iXjoXNP6MNL355CXKxwyWPLeHlVidchGQN4lCBUdZfzWAa8DJwElIpIHwDnsewwr52rqrmqmpuZmXlMceQ5EwTZALXx2ui+XXnlpmnkZHXn5udW8+s3NtBsN9UZjwU9QYhIqoiktT4HzgLWAa8AVzm7XQX8y+1Y8oorSUuKY3ivNLebMuaI0lMTmH/tyVw1eSBzFxZw9Z8/pqqu0euwTBTz4gyiN7BYRFYDHwOvq+pbwD3ADBHZAsxw1l2VX1xBTpYV6DOhIz42hrtmHs89F53A8oK9zHxkMVvLrNif8UbQy32ragEwtp3te4EzghVHVV0jm0v389UT+warSWP8NuukLIb17sL181dywSNL+f3Xx3Hm6N5eh2WiTChd5hpUhwr02fiDCVETBqbzyk1TGZSRynXz83jkg612U50JqqhNEHnFFcRagT4T4vp2T+aFGyYzc2xf7nt7Ezc9s4q6Biv2Z4IjehNEUSXH9+1KSoJNqmdCW1J8LA98fRx3nDuSN9ft5uJHl1mxPxMUUZkgGptbWF1iBfpM+BAR5pwyhKeunkhJZR0zH1nC8oK9XodlIlxUfn3+dFc1BxutQJ8JP18a0Yt/3TiV657O4/InPuLinP6HSsX07ppE766J9E5LontKPCJ2dZ45NlGZIPbU1JPRJdEK9JmwNDizCy/fOJWfvryOd9Z/RmU790okxMbQq2vioaTRK61NAmnd1jWJtMQ4SyTmsCScr4rIzc3VvLy8o3qtqtoHw0SEg43NlNfUU1p9kNJq57HmIGWtz6t9z2vamckuOT72ULLo3TWJ3mm+BPKf5JJEn25JJMXbZFqRRETy21TSPqyoPIMALDmYiJEUH8uA9BQGpKd0uF9tfRNlNZ8rLtcFAAAN4klEQVRPGr5k4ntcW7KPd6sPcrCx5XOvS0mI5Zczj+drE/q7+TZMCIraBGFMtElNjGNQYhyDMlIPu4+qUlPfRFmbs5HnVuzghy+sZtm2vdx9wRi78i+K2L+0MeYQEaFrUjxdk+IZ6tQoO39sXx56bwt/+GArq0v28cg3chhxnNUviwZReZmrMcZ/cbEx3HLWCOZfczL76ho5/+HFPLdiu93VHQUsQRhj/DJtWAZvfH8audk9+MlLa7n5uU/Y387At4kcliCMMX7rlZbE09eczC0zhvPK6l2c/4fFrN9V7XVYxiWWIIwxnRIbI3zvjGE8c90k9tc3ccEfl/DX5cXW5RSBLEEYY47KpME9eeP705k0uCd3/nMdN/19FTUHbYKjSGIJwhhz1DK6JPKXqyfy47NH8Na6zzjvD4tZW1LldVgmQCxBGGOOSUyM8J0vDeXZOZNoaGrh4keX8pclhdblFAEsQRhjAmJidjqvf28604Zl8ItX1/Ptv66k6oB1OYUzSxDGmIBJT03giStzuePckSzYUMpXHlrEJzv2eR2WOUqWIIwxARUT45u74vkbJqMKl/xpKU8sKrAupzBkCcIY44qcrB688b3pfGlEL371+gauezqPfXUNXodlOsEShDHGNd1S4pl7xQR+dt5oPtxczrkPLiK/uMLrsIyfLEEYY1wlIlwzbRAv3jCF2Fjh0seW86cPt9HSYl1Ooc4ShDEmKMYO6M5r353OWaN7c8+bG7lm3gr27q/3OizTAUsQxpig6ZYczx+/mcPdM8ewdOtezn1oER8XWpdTqLL5IIwxQSUiXDE5m/FZPbjpmZXMmruMW2YM59wT+rjedpekODK7JNqMkn6K2jmpjTHeqznYyB0vr+PV1buC1mZyfCxZzhStA3v6lgHpKQxMT6F/jxQS4iK/Y8XmpDbGhLy0pHgemjWOWRMHsCcI4xFVBxop3ltH8d46dlTUsXhr+efm4I4R6NMtmaz0NomjZwoD01PJ6plCt+R412MMJZYgjDGeEhGmDs3wpG1Vpbymnu0VvqRRXFHH9r21bK+oY8GGUvbs//x9G92S4xnYM4Ws9JRDSSTLSR59uiYRExNZXVeWIIwxUUtE6NU1iV5dk8jNTv+vn++vb2KHkzy2V9Q6j3Ws3VnFW+s+o6nNpboJsTH0T09mcEYqkwb3ZNqwDEb0Tgvr8Q5LEMYYcxhdEuMY1acro/p0/a+fNTW3sLvqoHPm4Tvr2L63jk2lNSzYUAZAr7REpg3NYPrwDKYOzaBXWlKw38IxsQRhjDFHIS42hgHOYPc0Pt9FtmvfARZv2cOirXv4YFMZ/1i1E4CRx6UxfVgG04ZlclJ2OskJsV6E7je7iskYY1zU0qKs313Noi17WLSlnLyiShqaW0iIi2Fidg+mD8tk2tAMRvfpGrQxDH+vYrIEYYwxQXSgoZmPiypYtLmcxVv3sPGzGgB6piYwdWgG04ZlMH1YBn26JbsWQ9he5ioiZwMPArHAE6p6j8chGWNMwCQnxHLq8ExOHZ4JQFn1QRZv3eOcYezhFeeekKG9ujBtaAanDM/g5EE9SU0M/p/rkDqDEJFYYDMwAygBVgCXqer69va3MwhjTCRRVTaV1rBos2/84uPCvRxsbCE+Vhif1YNTnPGLE/p1I/YYuqPCsotJRCYDv1DVLzvrtwOo6m/a298ShDEmkh1sbCa/uPLQ+MWnu6oB3/0Y3z19KLOnDz6q44ZrF1M/YEeb9RLgZI9iMcYYTyXFxzJ1qO8S2dvOGcne/fUs2baXRZvL6d3V/UtmQy1BtHfO9LlTHBGZA8wByMrKCkZMxhgTEnp2SeT8sX05f2zfoLQXalWpSoABbdb7A5+r4qWqc1U1V1VzMzMzgxqcMcZEk1BLECuAYSIySEQSgFnAKx7HZIwxUSmkuphUtUlEbgLexneZ61Oq+qnHYRljTFQKqQQBoKpvAG94HYcxxkS7UOtiMsYYEyIsQRhjjGmXJQhjjDHtsgRhjDGmXSFVaqOzRKQcKD7Kl2cAewIYjpft2HsJvTaC1U6ktBGsdiKljWNtZ6CqHvFGsrBOEMdCRPL8qUUSDu3Yewm9NoLVTqS0Eax2IqWNYLVjXUzGGGPaZQnCGGNMu6I5QcyNoHbsvYReG8FqJ1LaCFY7kdJGUNqJ2jEIY4wxHYvmMwhjjDEdiLoEISJPiUiZiKxzsY0BIvKBiGwQkU9F5PsutZMkIh+LyGqnnbvcaMdpK1ZEVonIay4dv0hE1orIJyLi2jSBItJdRF4UkY3Ov8/kAB9/hPMeWpdqEflBINtw2rnZ+TdfJyJ/FxFXZo8Rke87bXwayPfR3udQRNJF5F0R2eI89nChjUuc99IiIsd8BdBh2rjP+f+1RkReFpHuLrVzt9PGJyLyjogEfpIIVY2qBTgFyAHWudhGHyDHeZ6Gb57t0S60I0AX53k88BEwyaX3dAvwDPCaS8cvAjKC8O8/D5jtPE8AurvYVizwGb5rzgN53H5AIZDsrD8PXO1C/McD64AUfIU9FwDDAnTs//ocAvcCtznPbwN+60Ibo4ARwL+BXJfex1lAnPP8t8f6Pjpop2ub598D/hTo/wNRdwahqguBCpfb2K2qK53nNcAGfB/qQLejqrrfWY13loAPKolIf+ArwBOBPnYwiUhXfB+0JwFUtUFV97nY5BnANlU92ps5OxIHJItIHL4/4LuOsP/RGAUsV9U6VW0CPgQuDMSBD/M5nIkvgeM8XhDoNlR1g6puOpbj+tHGO87vC2A5vonP3Ginus1qKi589qMuQQSbiGQD4/F9u3fj+LEi8glQBryrqm6083vgx0CLC8dupcA7IpLvTCvrhsFAOfBnp7vsCRFJdakt8E149fdAH1RVdwL/B2wHdgNVqvpOoNvBd/Zwioj0FJEU4Fw+P+NjoPVW1d3g+5IF9HKxrWC5BnjTrYOLyP+KyA7gm8DPAn18SxAuEpEuwEvAD76Q7QNGVZtVdRy+byknicjxgTy+iJwHlKlqfiCP246pqpoDnAPcKCKnuNBGHL7T9EdVdTxQi68rI+CcGRHPB15w4dg98H3bHgT0BVJF5PJAt6OqG/B1kbwLvAWsBpo6fJE5RER+iu/39Te32lDVn6rqAKeNmwJ9fEsQLhGReHzJ4W+q+g+323O6Sv4NnB3gQ08FzheRIuBZ4HQR+WuA20BVdzmPZcDLwEmBbgPfnOclbc6yXsSXMNxwDrBSVUtdOPaZQKGqlqtqI/APYIoL7aCqT6pqjqqegq+LY4sb7ThKRaQPgPNY5mJbrhKRq4DzgG+qM0jgsmeAiwN9UEsQLhARwdfPvUFVf+diO5mtV0iISDK+PxwbA9mGqt6uqv1VNRtfl8n7qhrQb6sikioiaa3P8Q3yBfwqM1X9DNghIiOcTWcA6wPdjuMyXOhecmwHJolIivN/7Qx841wBJyK9nMcs4CLce0/gm3/+Kuf5VcC/XGzLNSJyNvAT4HxVrXOxnWFtVs8nwJ99ICqvYvo7vn7bRnzfKK91oY1p+PrU1wCfOMu5LrRzIrDKaWcd8DOXf3dfwoWrmPCNDax2lk+Bn7r4HsYBec7v7J9ADxfaSAH2At1cfB93OX8Q1gHzgUSX2lmEL4muBs4I4HH/63MI9ATew3eW8h6Q7kIbFzrP64FS4G0X2tgK7Gjz2T/mq4sO085Lzr//GuBVoF+g//3tTmpjjDHtsi4mY4wx7bIEYYwxpl2WIIwxxrTLEoQxxph2WYIwxhjTLksQJmKJyANtK5CKyNsi8kSb9ftF5JZjOP4vROSH/m4PFBHJFpFvtFm/WkQedqs9E70sQZhIthTnDmMRiQEygDFtfj4FWOLPgUQkNuDRHb1s4BtH2smYY2UJwkSyJfynBMUYfDcV1YhIDxFJxFetdJX43OfMe7BWRL4OICJfEt+8Hs8Aa51tPxWRTSKyAF/ZaL+JyOXim7/jExF5rDXpiMh+p+jaahFZLiK9ne1DnPUVIvJLEWmt3HsPMN05zs3Otr4i8pYzl8K9R/0bM6YNSxAmYqmvvlOTUyZiCrAMX1XdyUAusEZVG/CVkBgHjMVXruS+1ppA+GpC/VRVR4vIBHzlRsY7r5nobywiMgr4Or6ihOOAZnwVOMFXqnm5qo4FFgLXOdsfBB5U1Yl8vpz3bcAiVR2nqg8428Y5xz8B+LqIuFl11UQJSxAm0rWeRbQmiGVt1pc6+0wD/q6+yril+OY9aP3j/7GqFjrPpwMvq29+hGp8tYP8dQYwAVjhlGc/A1+JEYAGoHWmvnx8XUjgS2St1WCfOcLx31PVKlU9iK80xsBOxGZMu+K8DsAYl7WOQ5yAr4tpB3ArUA085ewjHby+9gvrR1ubRoB5qnp7Oz9r1P/UvGnm6D6X9W2eH+0xjPkcO4MwkW4JvrLLFc4ZQgXQHd+382XOPgvxdcvEikgmvlnnPm7nWAuBC0Uk2ak++9VOxPEe8LU21VHTReRI3/KX858SzrPabK/BN5WtMa6yBGEi3Vp8Vy8t/8K2KlXd46y/jK8i5mrgfeDH6isN/jnqm0b2OXwVOl/CV+n0cO4UkZLWRVXXA3fimzVvDb5JePp08HqAHwC3iMjHzr5VzvY1+MZWVrcZpDYm4KyaqzEhypnm84CqqojMAi5T1Zlex2Wih/VTGhO6JgAPO5MC7cM3v7ExQWNnEMYYY9plYxDGGGPaZQnCGGNMuyxBGGOMaZclCGOMMe2yBGGMMaZdliCMMca06/8BxrQyvgjYTIYAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import nltk, matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "plt.plot([f[0] for f in orderedLengthFreqs], [f[1] for f in orderedLengthFreqs])\n",
    "plt.xticks(range(1,len(orderedLengthFreqs)+1))\n",
    "plt.xlabel('Word Length')\n",
    "plt.ylabel('Word Count')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's compare that to Mendenhall's graph – not bad!\n",
    "\n",
    "![Mendhall Characteristic Curve](data/Mendenhall-CharacteristicCurve/OliverTwist-CharacteristicCurve.png)\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For the sake of convenience, we might want to create a function that combines all of these steps in one."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "import nltk, matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "def plot_characteristic_curve(string, limit=False):\n",
    "    tokens = nltk.word_tokenize(string)\n",
    "    words = [token for token in tokens if any(c for c in token if c.isalpha())]\n",
    "    if limit:\n",
    "        words = words[:limit]\n",
    "    lengths = [len(word) for word in words]\n",
    "    lengthFreqs = nltk.FreqDist(lengths)\n",
    "    orderedLengthFreqs = sorted(list(lengthFreqs.items()))\n",
    "    plt.plot([f[0] for f in orderedLengthFreqs], [f[1] for f in orderedLengthFreqs])\n",
    "    plt.xticks(range(1,len(orderedLengthFreqs)+1))\n",
    "    plt.xlabel('Word Length')\n",
    "    plt.ylabel('Word Count')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can try this out by plotting all of _Oliver Twist_ (something that Mendenhall probably wouldn't have wanted to do manually)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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"
    }
   ],
   "source": [
    "plot_characteristic_curve(oliverTwistString) # show all, without limiting to first 1000 words"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "\n",
    "This notebook is part of the [Epistemelogica](/) project by Stéfan Sinclair and Geoffrey Rockwell (CC-BY)."
   ]
  }
 ],
 "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": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}