compile_statistics.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Compile statistics from text files\n",
    "\n",
    "The code below is used to capture a number of data points from the OCRd text files that can be used to detect patterns and classify the documents into bins that require different processing workflows.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Setup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import csv\n",
    "import os\n",
    "import regex as re # https://pypi.org/project/regex/\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from scipy.signal import savgol_filter\n",
    "import pandas as pd\n",
    "from IPython.core.interactiveshell import InteractiveShell\n",
    "from ipywidgets import IntProgress\n",
    "from IPython.display import display\n",
    "\n",
    "plt.style.use(['dark_background'])\n",
    "InteractiveShell.ast_node_interactivity = \"none\"\n",
    "pd.options.display.max_rows = 100\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# progress bar\n",
    "def p_bar(mx, desc):\n",
    "    global p\n",
    "    p = IntProgress(min=0, max=mx, description=desc)\n",
    "    return p\n",
    "\n",
    "# paths\n",
    "english_dict_path = 'data/nltk_english_words'  # path to dictionary of all English words\n",
    "\n",
    "# production:\n",
    "raw_text_path = 'data/raw_text/'\n",
    "save_data_path = 'data/'\n",
    "\n",
    "# testing:\n",
    "#raw_text_path = 'test_data/sample/'\n",
    "#save_data_path = 'test_data/'\n",
    "\n",
    "# patterns \n",
    "ascii_string = set(\"\"\"!\"#$%&\\'()*+,-./0123456789:; =?ABCDEFGHIJKLMNOPQRSTUVWXYZ[\\\\]_`abcdefghijklmnopqrstuvwxyz\"\"\")\n",
    "word_chars = re.compile(\"[^a-zA-Z0-9 -']\")\n",
    "\n",
    "# load English dictionary\n",
    "with open(english_dict_path, 'r') as file:\n",
    "    english_words = set(file.read().split('\\n'))\n",
    "\n",
    "# create a list of all files in the folder\n",
    "raw_files = [f'{raw_text_path}{file.name}' for file in os.scandir(raw_text_path) if file.name != \".DS_Store\"]\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Extract statistics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5b8c4fd31e0a405ab1d99ca79ffe2f3c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "IntProgress(value=0, description='Processing', max=9838)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "file_stats = {}  # dictionary to store the results\n",
    "display(p_bar(len(raw_files), 'Processing')) # show the progress bar\n",
    "\n",
    "for i, fpath in enumerate(raw_files):\n",
    "    with open(fpath, 'r') as file:\n",
    "        text = file.read()\n",
    "    \n",
    "    # determine presence of \"JOINT RESOLUTION\" string\n",
    "    # fuzzy match parmitting 3 errors, ie. subs|dels|ins\n",
    "    jt_matches = re.search('(JOINT RESOLUTION){e<=3}', text)\n",
    "    \n",
    "    line_count = len(text.split('\\n')) # number of lines\n",
    "    char_count = len(text)  # number of characters\n",
    "    ascii_count = sum(c in ascii_string for c in text)  # percent of ascii characters\n",
    "    \n",
    "    text = word_chars.sub('', text) # clean non-word characters\n",
    "    tokens = [w.lower() for w in text.split(' ')]  # tokenize\n",
    "    word_count = len(tokens)  # number of words\n",
    "    english_count = sum(c in english_words for c in tokens)  # percent of english words\n",
    "    \n",
    "    # add metrics to dict\n",
    "    filename = os.path.basename(fpath)  # get filename\n",
    "    file_stats[filename] = {\n",
    "        'lines': line_count,\n",
    "        'chars': char_count,\n",
    "        'words': ascii_count,\n",
    "        'ascii_chars': round(ascii_count*100/char_count, 2),\n",
    "        'english_words': round(english_count*100/word_count, 2),\n",
    "        'jr_string': 1 if jt_matches is not None else 0\n",
    "    }\n",
    "    \n",
    "    if i % 100 == 0:\n",
    "        p.value += 100  #update progress bar\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>lines</th>\n",
       "      <th>chars</th>\n",
       "      <th>words</th>\n",
       "      <th>ascii_chars</th>\n",
       "      <th>english_words</th>\n",
       "      <th>jt_string</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>91-hjres1029-000ih-19691209.txt</th>\n",
       "      <td>27</td>\n",
       "      <td>1269</td>\n",
       "      <td>1218</td>\n",
       "      <td>95.98</td>\n",
       "      <td>70.11</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>94-hjres272-000ih-19750304.txt</th>\n",
       "      <td>28</td>\n",
       "      <td>1180</td>\n",
       "      <td>1132</td>\n",
       "      <td>95.93</td>\n",
       "      <td>70.55</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90-hjres1137-000ih-19680227.txt</th>\n",
       "      <td>25</td>\n",
       "      <td>1076</td>\n",
       "      <td>1034</td>\n",
       "      <td>96.10</td>\n",
       "      <td>72.33</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>78-hjres318-000ih-19441128.txt</th>\n",
       "      <td>40</td>\n",
       "      <td>1278</td>\n",
       "      <td>1218</td>\n",
       "      <td>95.31</td>\n",
       "      <td>70.86</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>92-sjres34-000is-19710126.txt</th>\n",
       "      <td>40</td>\n",
       "      <td>1990</td>\n",
       "      <td>1936</td>\n",
       "      <td>97.29</td>\n",
       "      <td>60.36</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>101-sjres9-000is-19890125.txt</th>\n",
       "      <td>68</td>\n",
       "      <td>2220</td>\n",
       "      <td>2133</td>\n",
       "      <td>96.08</td>\n",
       "      <td>71.70</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bills-106hjres3ih.txt</th>\n",
       "      <td>32</td>\n",
       "      <td>1164</td>\n",
       "      <td>1115</td>\n",
       "      <td>95.79</td>\n",
       "      <td>68.42</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>88-hjres28-000ih-19630109.txt</th>\n",
       "      <td>209</td>\n",
       "      <td>6057</td>\n",
       "      <td>5814</td>\n",
       "      <td>95.99</td>\n",
       "      <td>78.44</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bills-104hjres11ih.txt</th>\n",
       "      <td>159</td>\n",
       "      <td>4546</td>\n",
       "      <td>4361</td>\n",
       "      <td>95.93</td>\n",
       "      <td>74.81</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>76-hjres270-000ih-19390419.txt</th>\n",
       "      <td>23</td>\n",
       "      <td>901</td>\n",
       "      <td>865</td>\n",
       "      <td>96.00</td>\n",
       "      <td>68.55</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>9838 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 lines  chars  words  ascii_chars  \\\n",
       "91-hjres1029-000ih-19691209.txt     27   1269   1218        95.98   \n",
       "94-hjres272-000ih-19750304.txt      28   1180   1132        95.93   \n",
       "90-hjres1137-000ih-19680227.txt     25   1076   1034        96.10   \n",
       "78-hjres318-000ih-19441128.txt      40   1278   1218        95.31   \n",
       "92-sjres34-000is-19710126.txt       40   1990   1936        97.29   \n",
       "...                                ...    ...    ...          ...   \n",
       "101-sjres9-000is-19890125.txt       68   2220   2133        96.08   \n",
       "bills-106hjres3ih.txt               32   1164   1115        95.79   \n",
       "88-hjres28-000ih-19630109.txt      209   6057   5814        95.99   \n",
       "bills-104hjres11ih.txt             159   4546   4361        95.93   \n",
       "76-hjres270-000ih-19390419.txt      23    901    865        96.00   \n",
       "\n",
       "                                 english_words  jt_string  \n",
       "91-hjres1029-000ih-19691209.txt          70.11          0  \n",
       "94-hjres272-000ih-19750304.txt           70.55          1  \n",
       "90-hjres1137-000ih-19680227.txt          72.33          1  \n",
       "78-hjres318-000ih-19441128.txt           70.86          1  \n",
       "92-sjres34-000is-19710126.txt            60.36          0  \n",
       "...                                        ...        ...  \n",
       "101-sjres9-000is-19890125.txt            71.70          1  \n",
       "bills-106hjres3ih.txt                    68.42          1  \n",
       "88-hjres28-000ih-19630109.txt            78.44          1  \n",
       "bills-104hjres11ih.txt                   74.81          1  \n",
       "76-hjres270-000ih-19390419.txt           68.55          1  \n",
       "\n",
       "[9838 rows x 6 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "file_stats_df = pd.DataFrame.from_dict(file_stats, orient='index')  # create a dataframe with the results\n",
    "\n",
    "# save the results to a CSV file\n",
    "file_stats_df.to_csv(f'{save_data_path}text_file_statistics.csv', header=[\n",
    "    'filename', \n",
    "    'chars', \n",
    "    'words',\n",
    "    'ascii_chars',\n",
    "    'english_words',\n",
    "    'jr_string'\n",
    "])\n",
    "\n",
    "display(file_stats_df)  # display the results\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>lines</th>\n",
       "      <th>chars</th>\n",
       "      <th>words</th>\n",
       "      <th>ascii_chars</th>\n",
       "      <th>english_words</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>9838.000000</td>\n",
       "      <td>9.838000e+03</td>\n",
       "      <td>9.838000e+03</td>\n",
       "      <td>9838.000000</td>\n",
       "      <td>9838.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>133.252185</td>\n",
       "      <td>4.038591e+03</td>\n",
       "      <td>3.860114e+03</td>\n",
       "      <td>95.624264</td>\n",
       "      <td>72.176338</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1860.356820</td>\n",
       "      <td>4.956045e+04</td>\n",
       "      <td>4.723138e+04</td>\n",
       "      <td>1.689870</td>\n",
       "      <td>4.865600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>15.710000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>33.000000</td>\n",
       "      <td>1.309000e+03</td>\n",
       "      <td>1.253000e+03</td>\n",
       "      <td>95.550000</td>\n",
       "      <td>70.470000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>49.000000</td>\n",
       "      <td>1.578000e+03</td>\n",
       "      <td>1.510000e+03</td>\n",
       "      <td>95.950000</td>\n",
       "      <td>72.680000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>74.000000</td>\n",
       "      <td>2.325000e+03</td>\n",
       "      <td>2.222750e+03</td>\n",
       "      <td>96.230000</td>\n",
       "      <td>75.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>109758.000000</td>\n",
       "      <td>2.764147e+06</td>\n",
       "      <td>2.638264e+06</td>\n",
       "      <td>97.850000</td>\n",
       "      <td>100.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               lines         chars         words  ascii_chars  english_words\n",
       "count    9838.000000  9.838000e+03  9.838000e+03  9838.000000    9838.000000\n",
       "mean      133.252185  4.038591e+03  3.860114e+03    95.624264      72.176338\n",
       "std      1860.356820  4.956045e+04  4.723138e+04     1.689870       4.865600\n",
       "min         1.000000  1.000000e+00  0.000000e+00     0.000000      15.710000\n",
       "25%        33.000000  1.309000e+03  1.253000e+03    95.550000      70.470000\n",
       "50%        49.000000  1.578000e+03  1.510000e+03    95.950000      72.680000\n",
       "75%        74.000000  2.325000e+03  2.222750e+03    96.230000      75.000000\n",
       "max    109758.000000  2.764147e+06  2.638264e+06    97.850000     100.000000"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# show descriptive stats for key metrics\n",
    "display(file_stats_df.describe()[\n",
    "    ['lines', 'chars', 'words', 'ascii_chars', 'english_words']\n",
    "])\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The mean for ```english_words``` is 72.2 while the ```min``` is 15.7, the outliers should correspond to the handwritten documents where the OCR engine returns just gibberish. We can confirm with a quick chart:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(12, 5))\n",
    "plt.hist(file_stats_df['english_words'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1080 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axs = plt.subplots(2, 3, figsize=(20, 15))\n",
    "axs[0, 0].boxplot(file_stats_df['lines'])\n",
    "axs[0, 1].boxplot(file_stats_df['words'])\n",
    "axs[0, 2].boxplot(file_stats_df['chars'])\n",
    "axs[1, 0].boxplot(file_stats_df['english_words'])\n",
    "axs[1, 1].boxplot(file_stats_df['ascii_chars'])\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}