5.paper-analysis.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Notebook to replicate our paper's quantitative analysis\n",
    "\n",
    "- Author: Hadi Asghari & Freya Hewett\n",
    "- Version: 2023.02\n",
    "\n",
    "The first part of this notebook takes as input the classified Oscar22 DE-subset pages.\n",
    "\n",
    "The second part uses the LS-scraper results. It also makes use of Language Tool's API (explained further below). \n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "start_time": "2023-04-21T16:36:22.862490Z",
     "end_time": "2023-04-21T16:36:23.988827Z"
    }
   },
   "outputs": [],
   "source": [
    "from collections import Counter\n",
    "import os\n",
    "import requests\n",
    "import json\n",
    "import pandas as pd\n",
    "from scipy import stats\n",
    "import numpy as np\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Part I: analysis of web classification per domain category"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "ExecuteTime": {
     "start_time": "2023-04-21T17:13:54.802642Z",
     "end_time": "2023-04-21T17:14:37.818474Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16658597\n"
     ]
    },
    {
     "data": {
      "text/plain": "                                                 url               domain  \\\n0  http://cvd.bundesregierung.de/cvd-de/datenschu...   bundesregierung.de   \n1  http://demo.feuerwehr-kernen.de/fahrzeuge/eins...  feuerwehr-kernen.de   \n2  http://dl.ub.uni-freiburg.de/diglit/bibl_freim...      uni-freiburg.de   \n\n       domcat  readability  easy_score   clen  \n0  Government         11.3       0.000  31200  \n1     Society         13.3       0.000   7318  \n2   Education         13.3       0.097    202  ",
      "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>url</th>\n      <th>domain</th>\n      <th>domcat</th>\n      <th>readability</th>\n      <th>easy_score</th>\n      <th>clen</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>http://cvd.bundesregierung.de/cvd-de/datenschu...</td>\n      <td>bundesregierung.de</td>\n      <td>Government</td>\n      <td>11.3</td>\n      <td>0.000</td>\n      <td>31200</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>http://demo.feuerwehr-kernen.de/fahrzeuge/eins...</td>\n      <td>feuerwehr-kernen.de</td>\n      <td>Society</td>\n      <td>13.3</td>\n      <td>0.000</td>\n      <td>7318</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>http://dl.ub.uni-freiburg.de/diglit/bibl_freim...</td>\n      <td>uni-freiburg.de</td>\n      <td>Education</td>\n      <td>13.3</td>\n      <td>0.097</td>\n      <td>202</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# LOAD DATA\n",
    "df = pd.read_csv(\"./data/oscar22-classified.csv.xz\", sep='\\t')\n",
    "\n",
    "# or if recreating: \n",
    "# df = pd.read_csv(\"oscar-classified.csv\", sep='\\t', \n",
    "#                  names=['url', 'domain', 'domcat', 'readability', 'easy_score', 'clen'])\n",
    "# df = df[df.domain.str.endswith('de')]   \n",
    "\n",
    "print(len(df))\n",
    "df.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "ExecuteTime": {
     "start_time": "2023-04-21T17:14:37.818227Z",
     "end_time": "2023-04-21T17:14:39.209291Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "SignificanceResult(statistic=-0.4196474128263441, pvalue=0.0)"
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# spearman rho correlation between:\n",
    "# - the wiener_sachtextformel readibility score \n",
    "# - and our easy german classifier score\n",
    "# for cases where the readibility score is valid (among 5m randomly selected)\n",
    "t = df[(1<df.readability)&(df.readability<20)]\n",
    "stats.spearmanr(t.readability, t.easy_score)\n",
    "\n",
    "# SignificanceResult(statistic=-0.4196474128263441, pvalue=0.0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "ExecuteTime": {
     "start_time": "2023-04-21T17:14:39.214431Z",
     "end_time": "2023-04-21T17:14:51.227160Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                url  domain  perc_easy_u\n",
      "domcat                                  \n",
      "Science     2859859    7483     0.043210\n",
      "Gov/Region  2617151   24467     0.071892\n",
      "Business    1738920   36666     0.094679\n",
      "Shopping     642924    4812     0.181794\n",
      "Health       700874   12138     0.186790\n",
      "Society     1471080   15748     0.189874\n",
      "Computers    763282    3841     0.190214\n",
      "Recreation  2805087   28243     0.199405\n",
      "Travel       470097    9290     0.215796\n",
      "Kids         182729     864     0.248570\n",
      "Games        267319    1413     0.294192\n",
      "News        2139275    1236     0.305110\n"
     ]
    }
   ],
   "source": [
    "# Calculate percentage of 'easy' URLs/Domains (per category)\n",
    "\n",
    "# add columns\n",
    "df['is_easy'] = df['easy_score'] >= 0.8  \n",
    "#df['url_leicht'] = df.url.str.contains('leicht.+sprach', case=False)\n",
    "# let's combine a few categories for easier presentation\n",
    "df.loc[df.domcat=='Government', 'domcat'] = 'Gov/Region'\n",
    "df.loc[df.domcat=='RegOther', 'domcat'] = 'Gov/Region'\n",
    "df.loc[df.domcat=='Arts', 'domcat'] = 'Recreation'\n",
    "df.loc[df.domcat=='Education', 'domcat'] = 'Science'\n",
    "df.loc[df.domcat=='Home', 'domcat'] = 'Recreation'\n",
    "\n",
    "# URLs & domains per category\n",
    "tb2 = df.groupby('domcat').url.count().to_frame()\n",
    "tb2['domain'] = df.groupby('domcat').domain.nunique()\n",
    "\n",
    "# counts of only easy pages (and their domains)\n",
    "tb2['url_easy'] = df[df.is_easy].groupby('domcat').url.count()\n",
    "#tb['domain_weasy'] = df[df.is_easy].groupby('domcat').domain.nunique()\n",
    "tb2['perc_easy_u'] = tb2['url_easy'] / tb2['url']\n",
    "del tb2['url_easy']\n",
    "\n",
    "# print\n",
    "# one interesting finding is that so little of government urls is basic :)\n",
    "print(tb2.sort_values('perc_easy_u'))\n",
    "\n",
    "#  url  domain  perc_easy_u\n",
    "# domcat\n",
    "# Science        2859859    7483     0.043210\n",
    "# Gov_or_Region  2617151   24467     0.071892\n",
    "# Business       1738920   36666     0.094679\n",
    "# Shopping        642924    4812     0.181794\n",
    "# Health          700874   12138     0.186790\n",
    "# Society        1471080   15748     0.189874\n",
    "# Computers       763282    3841     0.190214\n",
    "# Recreation     2805087   28243     0.199405\n",
    "# Travel          470097    9290     0.215796\n",
    "# Kids            182729     864     0.248570\n",
    "# Games           267319    1413     0.294192\n",
    "# News           2139275    1236     0.305110"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\\begin{tabular}{lrrr}\n",
      "\\toprule\n",
      "{} &      url &  domain &  perc\\_easy\\_u \\\\\n",
      "domcat        &          &         &              \\\\\n",
      "\\midrule\n",
      "Science       &  2859859 &    7483 &     0.043191 \\\\\n",
      "Gov\\_or\\_Region &  2617151 &   24467 &     0.071788 \\\\\n",
      "Business      &  1738920 &   36666 &     0.094636 \\\\\n",
      "Shopping      &   642924 &    4812 &     0.181726 \\\\\n",
      "Health        &   700874 &   12138 &     0.186744 \\\\\n",
      "Society       &  1471080 &   15748 &     0.189807 \\\\\n",
      "Computers     &   763282 &    3841 &     0.190128 \\\\\n",
      "Recreation    &  2805087 &   28243 &     0.199335 \\\\\n",
      "Travel        &   470097 &    9290 &     0.215698 \\\\\n",
      "Kids          &   182729 &     864 &     0.248548 \\\\\n",
      "Games         &   267319 &    1413 &     0.294106 \\\\\n",
      "News          &  2139275 &    1236 &     0.305004 \\\\\n",
      "\\bottomrule\n",
      "\\end{tabular}\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_67349/58578050.py:1: FutureWarning: In future versions `DataFrame.to_latex` is expected to utilise the base implementation of `Styler.to_latex` for formatting and rendering. The arguments signature may therefore change. It is recommended instead to use `DataFrame.style.to_latex` which also contains additional functionality.\n",
      "  print(tb2.sort_values('perc_easy_u').to_latex())\n"
     ]
    }
   ],
   "source": [
    "# output for latex\n",
    "print(tb2.sort_values('perc_easy_u').to_latex())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "start_time": "2023-04-21T16:45:23.341534Z",
     "end_time": "2023-04-21T16:45:23.568252Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.15524968879432044\n"
     ]
    },
    {
     "data": {
      "text/plain": "<Axes: xlabel='domcat'>"
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": "<Figure size 640x480 with 1 Axes>",
      "image/png": "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\n"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# total stats & plot\n",
    "print(sum(df.is_easy)/len(df))\n",
    "tb2.get(['perc_easy_u']).plot(kind='bar')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "outputs": [
    {
     "data": {
      "text/plain": "<Figure size 640x480 with 2 Axes>",
      "image/png": "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\n"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Additional graphical representations of T1\n",
    "# Bar chart with primary and secondary y-axes (stacked) plus a line chart?\n",
    "\n",
    "# Version of above with an additional stacked barchart added for num-of-easy-pages\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.ticker as mticker\n",
    "\n",
    "# Calculate num_easy_pages\n",
    "tb2['num_easy_pages'] = tb2['url'] * tb2['perc_easy_u']\n",
    "\n",
    "# Sort the data by % easy pages\n",
    "data = tb2.sort_values(by='perc_easy_u')\n",
    "\n",
    "fig, ax1 = plt.subplots()\n",
    "\n",
    "# Bar chart (num_of_pages)\n",
    "ax1.bar(data.index, data.url, color='b', alpha=0.6, label='Num. of web pages')\n",
    "\n",
    "# Bar chart (num_easy_pages)\n",
    "ax1.bar(data.index, data.num_easy_pages, color='orange', alpha=0.6, label='Num. of easy pages')\n",
    "\n",
    "ax1.set_xlabel('Website Type')\n",
    "#ax1.set_ylabel('N. of web pages', color='b')\n",
    "ax1.tick_params(axis='y', labelcolor='b')\n",
    "\n",
    "# Set y-axis limits and ticks for num_of_pages\n",
    "ax1.set_ylim(bottom=0)\n",
    "ax1.set_yticks([0, 1_000_000, 2_000_000, 3_000_000])\n",
    "ax1.set_yticklabels(['0', '1m', '2m', '3m'])\n",
    "\n",
    "# Line chart (% easy pages)\n",
    "ax2 = ax1.twinx()\n",
    "ax2.plot(data.index, np.array(data.perc_easy_u) * 100, 'r', marker='o', linestyle='--', label='Percentage easy')\n",
    "#ax2.set_ylabel('% easy pages', color='r')\n",
    "ax2.tick_params(axis='y', labelcolor='r')\n",
    "\n",
    "# Set y-axis limits and ticks for percent_easy_pages\n",
    "ax2.set_ylim(bottom=0)\n",
    "ax2.yaxis.set_major_formatter(mticker.FormatStrFormatter('%.0f%%'))\n",
    "ax2.set_yticks(range(0, 31, 5))\n",
    "\n",
    "# Tilt category names\n",
    "plt.setp(ax1.get_xticklabels(), rotation=45, ha='right')\n",
    "\n",
    "# Display legend\n",
    "lines, labels = ax1.get_legend_handles_labels()\n",
    "lines2, labels2 = ax2.get_legend_handles_labels()\n",
    "ax2.legend(lines + lines2, labels + labels2, loc='upper left')\n",
    "\n",
    "#plt.title('Num of Pages, Num of Easy Pages, and % Easy Pages by Website Type (Ordered by % Easy Pages)')\n",
    "plt.savefig('/home/hadi/lsweb-barchart2.png', dpi=300, bbox_inches='tight')\n",
    "\n",
    "plt.show()"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "start_time": "2023-04-21T17:29:32.931307Z",
     "end_time": "2023-04-21T17:29:33.558688Z"
    }
   }
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Part II: Analysis of LS web crawl\n",
    "\n",
    "We estimate the coverage/comprehensivenss of the Oscar (Common Crawl) by calculating what percentage of the LS URLs our crawler finds (which are among the public sector websites shared with us) exist in our Oscar subset.\n",
    "\n",
    "Due to anonimity of the states, we cannot share the list of found URLs. However, the results are as follows:\n",
    "\n",
    "- 11 of the public sector domains are missing in the subset—this is probably a curlie issue rather than Oscar\n",
    "- Among the remainder, out of 310 found URLs, only 9 appear in the subset. This is a crawler coverage issue\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# import tldextract\n",
    "# from urllib.parse import urlparse\n",
    "\n",
    "# lszero =  open(\"./data/found-lsurls-excluded.txt\").readlines()\n",
    "\n",
    "# if not 'urls' in locals():\n",
    "#     urls = set(df.url)\n",
    "#     domains = set(df.domain)\n",
    "\n",
    "# u_in, u_out = [], []\n",
    "# d_in, d_out = set(), set()\n",
    "\n",
    "# for u in lszero:\n",
    "#     u = u.strip()\n",
    "#     d = tldextract.extract(urlparse(u).netloc).registered_domain\n",
    "#     if d in domains:\n",
    "#         d_in.add(d)\n",
    "#         if u in urls:\n",
    "#             u_in.append(u)\n",
    "#         else:\n",
    "#             u_out.append(u)\n",
    "#     else:\n",
    "#         d_out.add(d)\n",
    "\n",
    "\n",
    "# print('domains:', len(d_in), len(d_out))\n",
    "# print('urls:', len(u_in), len(u_out), )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Table 2**\n",
    "\n",
    "To calculate the number of domains that have LS pages (which are presented in Table 2), please refer to the results/output-log of the crawler.\n",
    "\n",
    "Regarding the LS quality aspect on the pages found, the following cell calculates the amount of times that a [Leichte Sprache rule](https://languagetool.org/de/leichte-sprache) is broken. The input web pages have been saved as text files by the web crawler. The rule check code uses the [Language Tool API](https://dev.languagetool.org/public-http-api), which we recommend [installing locally](https://dev.languagetool.org/http-server). \n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "inputdir = \"./crawler/savedpages/\"  # path where input files are saved\n",
    "outputfile = \"output_file.csv\" # path to output csv file\n",
    "bundesland = \"bundesland\" # first column in output csv file\n",
    "\n",
    "texts = dict()\n",
    "filenames = os.listdir(inputdir)\n",
    "for fn in filenames:\n",
    "    if fn.endswith('.txt'):\n",
    "        with open(inputdir + fn, 'r') as in_file:\n",
    "            text = in_file.readlines()\n",
    "            body = ' '.join(text[1:])\n",
    "            body = body.replace('\\n', '')\n",
    "            # replace * and # because they count as sonderzeichen\n",
    "            body = body.replace('*', '')\n",
    "            body = body.replace('#', '')\n",
    "            texts[text[0].strip()] = body\n",
    "\n",
    "all_counters = dict()\n",
    "\n",
    "for i,t in enumerate(texts):\n",
    "    if i% 10 == 0:\n",
    "        print(str(i), \" texts parsed\")\n",
    "\n",
    "    # Get length of text (characters)\n",
    "    length = len(texts[t])\n",
    "\n",
    "    headers = {'Content-Type': 'application/x-www-form-urlencoded',}\n",
    "    data = 'text=' + texts[t] + '&language=de-DE-x-simple-language'\n",
    "\n",
    "    # We recommend setting up your own LanguageTool server for speed/reliability. \n",
    "    # If you are just trying out one text, you can use the public API: https://api.languagetool.org/v2/check\n",
    "    # (but please note the restrictions)\n",
    "    response = requests.post('http://localhost:8081/v2/check', headers=headers, data=data.encode('utf-8'))\n",
    "    print(response.status_code)\n",
    "    if response.status_code != 400:\n",
    "        output = response.json()\n",
    "    else:\n",
    "        print(i)\n",
    "        continue\n",
    "\n",
    "    rule_breaks = Counter()\n",
    "    for matches in output['matches']:\n",
    "        rule_breaks[matches['rule']['id']] += 1\n",
    "\n",
    "    #Total rule breaks\n",
    "    total_rule_breaks = sum(rule_breaks.values())\n",
    "\n",
    "    #Unique rule breaks\n",
    "    unique_rule_breaks = len(list(rule_breaks))\n",
    "\n",
    "    #Top 3 rule breaks\n",
    "    top_3 = rule_breaks.most_common(3)\n",
    "\n",
    "    all_counters[t] = rule_breaks\n",
    "\n",
    "    #CSV file with: URL, length of text in characters, total rule breaks, unique rule breaks,\n",
    "    with open(outputfile, 'a') as output:\n",
    "        print(bundesland+','+t+','+str(length)+','+str(total_rule_breaks)+','+str(unique_rule_breaks)+',\"'+str(top_3)+'\"',file=output)\n",
    "\n",
    "    json.dump(all_counters, open('counter_' + bundesland + '.json', 'w'))\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": "3.10.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}