sec-priv-tweets.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "cff32eeb-926b-4d0f-b0d0-67636deb0bae",
   "metadata": {},
   "outputs": [],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "from tqdm import tqdm\n",
    "import pandas as pd\n",
    "from matplotlib import pyplot as plt\n",
    "import numpy as np\n",
    "from collections import Counter\n",
    "import operator,json, re, itertools \n",
    "from urllib.parse import urlparse\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "# from wordcloud import WordCloud \n",
    "pd.options.display.max_colwidth = 300\n",
    "import seaborn as sns\n",
    "import config\n",
    "data_root= config.DATA_ROOT\n",
    "tqdm.pandas()\n",
    "# import collect_tweets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "f5dbc8cc-a590-4e65-875d-afac8d275165",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "# of privacy related tweets: 216667\n",
      "# of unique users who tweeted:90579\n"
     ]
    }
   ],
   "source": [
    "df=pd.read_pickle(data_root+'tweets-priv.pkl')\n",
    "print(\"# of privacy related tweets: {}\\n# of unique users who tweeted:{}\".format(len(df), len(df.author_id.unique())))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "2db3f7e3-f7d9-415b-aaca-b9d9c07ac4dc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "total # of users 1698096\n"
     ]
    }
   ],
   "source": [
    "user_df = pd.read_pickle(data_root+'users-en-unique.pkl')\n",
    "print('total # of users', len(user_df))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "41537421-f0e3-412a-909c-3a8672bf5402",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████| 216667/216667 [00:31<00:00, 6855.93it/s]\n"
     ]
    }
   ],
   "source": [
    "uids = set(user_df.index)\n",
    "df['author_group'] = df.progress_apply(lambda row: user_df.loc[row.author_id].user_group if row.author_id in uids else 'Unknown', axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "39474fdd-06c4-4559-81fa-60fd3d197d09",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "author_group\n",
       "personal            2254\n",
       "techie              2492\n",
       "academic            8197\n",
       "news               12314\n",
       "academic_techie    19170\n",
       "personal_brand     20846\n",
       "business_brand     44097\n",
       "Unknown            53009\n",
       "unknown            54288\n",
       "Name: text, dtype: int64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'''Number of tweets posted across groups'''\n",
    "df.groupby('author_group').text.count().sort_values(ascending=True)\n",
    "# print(df.groupby('author_group').text.count().sort_values(ascending=True).to_frame().to_latex())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "6ccf49c2-21d3-41b9-8d76-76229003b3bc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "# of users who tweeted sec-priv tweets but deleted their accounts: 26772\n",
      "# of users who tweeted sec-priv tweets and have their accounts: 63807\n"
     ]
    }
   ],
   "source": [
    "deleted_users = set(df.author_id.unique()).difference(user_df.index)\n",
    "print('# of users who tweeted sec-priv tweets but deleted their accounts:', len(deleted_users))\n",
    "current_users = set(df.author_id).intersection(user_df.index)\n",
    "print('# of users who tweeted sec-priv tweets and have their accounts:', len(current_users))\n",
    "user_df=user_df.loc[list(current_users)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "d8e1d028-e459-43c1-98ae-140c749000e6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(163658, 25)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'''remove tweets from deleted users as per TOS'''\n",
    "df=df[df.apply(lambda row: row.author_id in current_users, axis=1)]\n",
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "a5cff502-b914-40d3-b11c-7e5b6c11a4cf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "author_group\n",
       "personal            2254\n",
       "techie              2492\n",
       "academic            8197\n",
       "news               12314\n",
       "academic_techie    19170\n",
       "personal_brand     20846\n",
       "business_brand     44097\n",
       "unknown            54288\n",
       "Name: text, dtype: int64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'''Number of tweets posted across groups after removing deleted profiles'''\n",
    "df.groupby('author_group').text.count().sort_values(ascending=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "0c18c82f-bc1e-404c-9f6c-8456aa5fe1bc",
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"created_at\"] = pd.to_datetime(df[\"created_at\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "1aa9d2ce-94ab-4aa8-9a51-8ee17ad70083",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['search_grp'] = df.apply(lambda tweet: \n",
    "                            'mention' if tweet.search_term.startswith('@') \\\n",
    "                            else ('hashtag' if tweet.search_term.startswith('#') else 'other'), axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "fa8974f8-b8d2-4458-8e8f-37e6927e3f24",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 85%|██████████████████████████████████████████████████████████████████████████████████████████████                | 139904/163658 [00:06<00:00, 37854.45it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1470755533770694661 Expecting ',' delimiter: line 1 column 763 (char 762)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████| 163658/163658 [00:06<00:00, 24561.35it/s]\n"
     ]
    }
   ],
   "source": [
    "'''Parse the entities in a tweet object'''\n",
    "\n",
    "def singleQuoteToDoubleQuote(singleQuoted):\n",
    "        '''\n",
    "        convert a single quoted string to a double quoted one\n",
    "        Args:\n",
    "            singleQuoted(string): a single quoted string e.g. {'cities': [{'name': \"Upper Hell's Gate\"}]}\n",
    "        Returns:\n",
    "            string: the double quoted version of the string e.g. \n",
    "        see\n",
    "           - https://stackoverflow.com/questions/55600788/python-replace-single-quotes-with-double-quotes-but-leave-ones-within-double-q \n",
    "        '''\n",
    "        cList=list(singleQuoted)\n",
    "        inDouble=False;\n",
    "        inSingle=False;\n",
    "        for i,c in enumerate(cList):\n",
    "            #print (\"%d:%s %r %r\" %(i,c,inSingle,inDouble))\n",
    "            if c==\"'\":\n",
    "                if not inDouble:\n",
    "                    inSingle=not inSingle\n",
    "                    cList[i]='\"'\n",
    "            elif c=='\"':\n",
    "                inDouble=not inDouble\n",
    "        doubleQuoted=\"\".join(cList)    \n",
    "        return doubleQuoted\n",
    "    \n",
    "def parse_entity(row):\n",
    "    if isinstance(row.entities, str):\n",
    "        try:\n",
    "            return json.loads(singleQuoteToDoubleQuote(row.entities).replace(u'\\\\xa0', u' '))\n",
    "        except Exception as e:\n",
    "            print(row.name, e)\n",
    "            return {}\n",
    "    else: \n",
    "        return {}\n",
    "df['entities_parsed']= df.progress_apply(lambda row: parse_entity(row), axis=1)\n",
    "# parse_entity(df.loc['1470058531546148880'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "f472a15d-6f6a-48fd-8c7a-a71c3792f28d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(104257, {'annotations', 'cashtags', 'hashtags', 'mentions', 'urls'})"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "keys = df.apply(lambda row: list(row.entities_parsed.keys()), axis=1)\n",
    "keys=list(itertools.chain.from_iterable(keys.values))\n",
    "len(keys),set(keys)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "8a2ca73f-b516-4b8a-8e11-ef625cd1d414",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['urls'] = df.apply(lambda row: [urlparse(url['expanded_url']).netloc for url in row.entities_parsed['urls']] \\\n",
    "                      if 'urls' in row.entities_parsed else [], axis=1)\n",
    "df['hashtags'] = df.apply(lambda row: [hashtag['tag'] for hashtag in row.entities_parsed['hashtags']] \\\n",
    "                      if 'hashtags' in row.entities_parsed else [], axis=1)\n",
    "df['anno_type'] = df.apply(lambda row: [a['type'] for a in row.entities_parsed['annotations']] \\\n",
    "                      if 'annotations' in row.entities_parsed else [], axis=1)\n",
    "df['anno_text'] = df.apply(lambda row: [a['normalized_text'] for a in row.entities_parsed['annotations']] \\\n",
    "                      if 'annotations' in row.entities_parsed else [], axis=1)\n",
    "df['mentions'] = df.apply(lambda row: [m['username'] for m in row.entities_parsed['mentions']] \\\n",
    "                      if 'mentions' in row.entities_parsed else [], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "70299edb-a50d-47cb-bd8f-3db97195a454",
   "metadata": {},
   "outputs": [],
   "source": [
    "def entity_stat(df, common=5):\n",
    "    print('Number of tweets with url(s):', len(df[df.apply(lambda row: len(row.urls)>0, axis=1)]))\n",
    "    print('Number of tweets with hashtag(s):', len(df[df.apply(lambda row: len(row.hashtags)>0, axis=1)]))\n",
    "    print('Number of tweets with annotations(s):', len(df[df.apply(lambda row: len(row.anno_type)>0, axis=1)]))\n",
    "    print('Number of tweets with mention(s):', len(df[df.apply(lambda row: len(row.mentions)>0, axis=1)]))\n",
    "    print()\n",
    "    mentions = list(itertools.chain.from_iterable(df.mentions.values))\n",
    "    tags=list(itertools.chain.from_iterable(df.hashtags.values))\n",
    "    urls=list(itertools.chain.from_iterable(df.urls.values))\n",
    "    anno_type=list(itertools.chain.from_iterable(df.anno_type.values))\n",
    "    anno_text=list(itertools.chain.from_iterable(df.anno_text.values))\n",
    "    print('urls: {}, unique:{}, most_common:{}'.format(len(urls), len(set(urls)), Counter(urls).most_common(common)))\n",
    "    print('mentions: {}, unique:{}, most_common:{}'.format(len(mentions), len(set(mentions)), Counter(mentions).most_common(common)))\n",
    "    print('tags: {}, unique:{}, most_common:{}'.format(len(tags), len(set(tags)), Counter(tags).most_common(common)))\n",
    "    print('anno_type: {}, unique:{}, most_common:{}'.format(len(anno_type), len(set(anno_type)), Counter(anno_type).most_common(common)))\n",
    "    print('anno_text: {}, unique:{}, most_common:{}'.format(len(anno_text), len(set(anno_text)), Counter(anno_text).most_common(common)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "00c4a739-89f8-4402-bb7d-f41ff6c2016a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of tweets with url(s): 40541\n",
      "Number of tweets with hashtag(s): 41776\n",
      "Number of tweets with annotations(s): 6666\n",
      "Number of tweets with mention(s): 15245\n",
      "\n",
      "urls: 50868, unique:3020, most_common:[('twitter.com', 11127), ('bit.ly', 8833), ('ow.ly', 6447), ('buff.ly', 2516), ('hubs.ly', 1440), ('gag.gl', 897), ('stwb.co', 858), ('okt.to', 799), ('goo.gl', 772), ('lnkd.in', 593)]\n",
      "mentions: 22541, unique:7879, most_common:[('ZDNet', 226), ('EdTech_K12', 220), ('CIOonline', 206), ('InfosecurityMag', 205), ('ITProPortal', 175), ('NTTComSec_UK', 168), ('helpnetsecurity', 161), ('SearchSecurity', 160), ('CSOonline', 150), ('MdMisi', 131)]\n",
      "tags: 144506, unique:10328, most_common:[('BYOD', 33591), ('security', 12834), ('cybersecurity', 5280), ('byod', 4103), ('infosec', 4082), ('mobile', 3653), ('privacy', 1553), ('mobilesecurity', 1463), ('IoT', 1444), ('Cybersecurity', 1401)]\n",
      "anno_type: 8125, unique:5, most_common:[('Organization', 3343), ('Person', 1790), ('Product', 1606), ('Place', 1183), ('Other', 203)]\n",
      "anno_text: 8125, unique:2989, most_common:[('Android', 296), ('Google', 152), ('AOTMP Telecom Management Daily', 131), ('UK', 130), ('Apple', 115), ('MSPs', 107), ('US', 97), ('iOS', 97), ('IBM', 84), ('Proctorio', 80)]\n"
     ]
    }
   ],
   "source": [
    "entity_stat(df , common=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "1f276319-68c2-4f46-b909-a60effecef7e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['author_id', 'text', 'created_at', 'geo', 'retweets', 'replies',\n",
       "       'likes', 'quote_count', 'lang', 'conversation_id',\n",
       "       'context_annotations', 'entities', 'attachments', 'possibly_sensitive',\n",
       "       'withheld', 'reply_settings', 'source', 'search_term', 'senti', 'topic',\n",
       "       'text_clean_topic', 'text_clean_senti', 'topic_cos', 'topic_con',\n",
       "       'author_group', 'search_grp', 'entities_parsed', 'urls', 'hashtags',\n",
       "       'anno_type', 'anno_text', 'mentions'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# t=df[df.apply(lambda row: 'twitter.com' in row.urls[0] if row.urls else False, axis=1)]\n",
    "# pd.options.display.max_colwidth=390\n",
    "# t[['text']]\n",
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "c6ce8dce-db80-4b25-9c94-2ab097146350",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_tweet_type(tweet):\n",
    "    for url in tweet.urls:\n",
    "        if 'twitter.com' in url:\n",
    "            return 'quote'\n",
    "    if int(tweet.name)!=tweet.conversation_id:\n",
    "        return 'reply'\n",
    "    return 'novel'\n",
    "\n",
    "df['tweet_type'] = df.apply(lambda tweet: get_tweet_type(tweet), axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "909fe55b-ee28-4a94-9bf1-25b93cebe465",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tweet_type\n",
       "novel    134332\n",
       "quote     11023\n",
       "reply     18303\n",
       "Name: text, dtype: int64"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby('tweet_type').text.count()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c2a8bb94-4a5d-48b0-9624-cd82ea966fb0",
   "metadata": {},
   "source": [
    "df.to_pickle(data_root+'tweets-priv-processed.pkl')## Who tweeted? Find user groups"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "60f82434-c11b-45fe-be51-dfb788140675",
   "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></th>\n",
       "      <th>text</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>author_group</th>\n",
       "      <th>tweet_type</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>business_brand</th>\n",
       "      <th>novel</th>\n",
       "      <td>37210</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>personal_brand</th>\n",
       "      <th>novel</th>\n",
       "      <td>17838</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>academic_techie</th>\n",
       "      <th>novel</th>\n",
       "      <td>15710</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>news</th>\n",
       "      <th>novel</th>\n",
       "      <td>11315</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>academic</th>\n",
       "      <th>novel</th>\n",
       "      <td>6078</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>business_brand</th>\n",
       "      <th>quote</th>\n",
       "      <td>5697</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>academic_techie</th>\n",
       "      <th>reply</th>\n",
       "      <td>2913</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>academic</th>\n",
       "      <th>reply</th>\n",
       "      <td>2000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>techie</th>\n",
       "      <th>novel</th>\n",
       "      <td>1964</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>personal_brand</th>\n",
       "      <th>quote</th>\n",
       "      <td>1788</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             text\n",
       "author_group    tweet_type       \n",
       "business_brand  novel       37210\n",
       "personal_brand  novel       17838\n",
       "academic_techie novel       15710\n",
       "news            novel       11315\n",
       "academic        novel        6078\n",
       "business_brand  quote        5697\n",
       "academic_techie reply        2913\n",
       "academic        reply        2000\n",
       "techie          novel        1964\n",
       "personal_brand  quote        1788"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df.author_group!='unknown'].groupby(['author_group', 'tweet_type']).text.count().sort_values(ascending=False).to_frame().head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "id": "42a6a322-fb71-41d7-8e47-e1fc9674cd78",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df[df.author_group!='unknown'].groupby(['author_group', 'tweet_type']).text.count().unstack().plot(kind='barh', width=.8, figsize=(12,8))\n",
    "plt.yticks(fontsize=14)\n",
    "plt.xticks(fontsize=14)\n",
    "plt.xlabel('count', fontsize=16)\n",
    "plt.ylabel('author group', fontsize=16)\n",
    "plt.legend(title=\"Tweet type\", title_fontsize=15, fontsize=14, fancybox=True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "id": "b74211b5-9d30-45de-ac14-597488fb4a6b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(18,6))\n",
    "df.reset_index().set_index('created_at').groupby(pd.Grouper(freq='3M')).text.count().plot()\n",
    "plt.title('')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "05a7070f-bf41-413d-92b8-05091031635a",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.to_pickle(data_root+'tweets-priv-processed.pkl')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "46aac43e-7f17-4953-90d8-7e5e1532e62b",
   "metadata": {},
   "source": [
    "## Tweet sentiment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "c5b07109-fb5f-426a-9c86-ec02fb030d60",
   "metadata": {},
   "outputs": [],
   "source": [
    "# def remove_comp(senti):\n",
    "#     if 'compound' in senti:\n",
    "#         senti.pop('compound')\n",
    "#     return senti\n",
    "# df['senti2'] = df.apply(lambda tweet: remove_comp(tweet.senti), axis=1)\n",
    "# df['senti2'] = df.apply(lambda tweet: max(tweet.senti2.items(), key=operator.itemgetter(1))[0], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "09212d49-f527-4ec4-8bdc-84e31f7b64c9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "senti\n",
       "negative      7309\n",
       "neutral     148082\n",
       "positive      8267\n",
       "Name: text, dtype: int64"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby('senti').text.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "965b08cb-c0d6-4c58-96da-536e05f0e6b7",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "7a120e24-5271-4610-88c8-6e183fb86140",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tweet_type  senti   \n",
       "novel       negative      5651\n",
       "            neutral     121958\n",
       "            positive      6736\n",
       "quote       negative       193\n",
       "            neutral      10318\n",
       "            positive       499\n",
       "reply       negative      1465\n",
       "            neutral      15806\n",
       "            positive      1032\n",
       "Name: text, dtype: int64"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby(['tweet_type','senti']).text.count()#.sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "0292c504-0fa7-4f16-9f75-40091e49ade9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "author_group\n",
       "unknown           3545\n",
       "business_brand    1327\n",
       "academic_techi     823\n",
       "personal_brand     579\n",
       "news               451\n",
       "academic           413\n",
       "techie             171\n",
       "Name: text, dtype: int64"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df.senti=='negative'].groupby(['author_group']).text.count().sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "1a828b8a-228f-4deb-bcac-300c7e9170ea",
   "metadata": {},
   "outputs": [],
   "source": [
    "neg_df = df[df.senti=='negative']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c9a9283e-f5ca-454d-9a31-20bd6114bbda",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "b916cc49-cfa7-460d-bea9-27801d9be24a",
   "metadata": {},
   "outputs": [],
   "source": [
    "sample1 = pd.read_excel('priv-tweets-neg-sample1.xls')\n",
    "sample1.tweetid = sample1.tweetid.astype(str)\n",
    "sample1.set_index('tweetid', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "8819b23b-a440-4f03-9a3a-6124fd478bbb",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/q5/tcs66y_j6z1bcz8ffp7_1m4r0000gq/T/ipykernel_55137/1178873254.py:1: FutureWarning: Passing a set as an indexer is deprecated and will raise in a future version. Use a list instead.\n",
      "  df.loc[(set(sample1.index).intersection(set(df[df.senti=='negative'].index)))].sample(10)[['text']]\n"
     ]
    },
    {
     "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>text</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>tweetid</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>388933694498406400</th>\n",
       "      <td>No Recourse for Girls Harassed by Transgender Bully http://t.co/XI0FlCUhAb CO girls' privacy jeopardized by one gender confused moron=wrong</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25920117445</th>\n",
       "      <td>I'm truly horrified by the circumstances of this young Rutgers student's invasion of privacy and subsequent suicide.  http://bit.ly/bKaQBj</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27787390340833280</th>\n",
       "      <td>Teacher Punished for Reporting Violent 7-Year-Old: http://abcn.ws/eP17QU. Did she violate the student's privacy, or do the right thing?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1455170658371977226</th>\n",
       "      <td>@CheetahJim1 @ProFootballTalk Not saying it's right to violate one's privacy but where were y'all when Linda Tripp was violating Monica Lewinski?  People get violated by the system every day.  Your typical college rape of a student by a professor involves far more violation than poor millionaire...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1337484131928862727</th>\n",
       "      <td>This is why I get mad when edtech or tech platforms get applauded \\n\\nNo one EVER asks what the fuck is going on in schools with privacy \\n\\nAnd let me tell you it is a mess because half the staff don’t know and the other half are wannabe prison guards https://t.co/GRkU0cBfHH</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1122934665122320384</th>\n",
       "      <td>A student cleared of rape when his accuser’s text messages proved his innocence said this morning that rape suspects deserve the same right to privacy as victims https://t.co/rt13k0DI34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1025695617048301568</th>\n",
       "      <td>@BostonDotCom WTF? This is disgusting. She's a student that work hard to get into college and she gets target b/c she a person of color and they're worried about the \"privacy laws\" after this young woman got humiliated  for doing nothing wrong. Get the hell out here with that BS.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>180695154313805824</th>\n",
       "      <td>Dharun Ravi, ex Rutgers Univ student was found guilty on charges of invasion of privacy, but not guilty charges of bias intimidation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9293828456</th>\n",
       "      <td>Absolutely horrifying: \"School used student laptop webcams to spy on them at school and at home\" http://bit.ly/apfgRf #privacy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>892577006730133504</th>\n",
       "      <td>Posting a student's work is a gross violation of privacy. But it's much less gross than their preferred method of traveling. https://t.co/5SpkcmR5Qy</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                                                                                                                                                                                                                                                                            text\n",
       "tweetid                                                                                                                                                                                                                                                                                                                         \n",
       "388933694498406400                                                                                                                                                                   No Recourse for Girls Harassed by Transgender Bully http://t.co/XI0FlCUhAb CO girls' privacy jeopardized by one gender confused moron=wrong\n",
       "25920117445                                                                                                                                                                           I'm truly horrified by the circumstances of this young Rutgers student's invasion of privacy and subsequent suicide.  http://bit.ly/bKaQBj\n",
       "27787390340833280                                                                                                                                                                        Teacher Punished for Reporting Violent 7-Year-Old: http://abcn.ws/eP17QU. Did she violate the student's privacy, or do the right thing?\n",
       "1455170658371977226  @CheetahJim1 @ProFootballTalk Not saying it's right to violate one's privacy but where were y'all when Linda Tripp was violating Monica Lewinski?  People get violated by the system every day.  Your typical college rape of a student by a professor involves far more violation than poor millionaire...\n",
       "1337484131928862727                         This is why I get mad when edtech or tech platforms get applauded \\n\\nNo one EVER asks what the fuck is going on in schools with privacy \\n\\nAnd let me tell you it is a mess because half the staff don’t know and the other half are wannabe prison guards https://t.co/GRkU0cBfHH\n",
       "1122934665122320384                                                                                                                    A student cleared of rape when his accuser’s text messages proved his innocence said this morning that rape suspects deserve the same right to privacy as victims https://t.co/rt13k0DI34\n",
       "1025695617048301568                     @BostonDotCom WTF? This is disgusting. She's a student that work hard to get into college and she gets target b/c she a person of color and they're worried about the \"privacy laws\" after this young woman got humiliated  for doing nothing wrong. Get the hell out here with that BS.\n",
       "180695154313805824                                                                                                                                                                          Dharun Ravi, ex Rutgers Univ student was found guilty on charges of invasion of privacy, but not guilty charges of bias intimidation\n",
       "9293828456                                                                                                                                                                                        Absolutely horrifying: \"School used student laptop webcams to spy on them at school and at home\" http://bit.ly/apfgRf #privacy\n",
       "892577006730133504                                                                                                                                                          Posting a student's work is a gross violation of privacy. But it's much less gross than their preferred method of traveling. https://t.co/5SpkcmR5Qy"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[(set(sample1.index).intersection(set(df[df.senti=='negative'].index)))].sample(10)[['text']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "b318d626-7a74-4937-a1de-eebf2ae110ad",
   "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>text</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>tweetid</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>101695842209107968</th>\n",
       "      <td>Court sides with college accused of snooping on student’s email account http://e74.us/8w #privacy RT @internetcases &amp;lt; Unencrypted = insecure</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>344785106378043393</th>\n",
       "      <td>BYOD &amp;amp; the implications for the IT team http://t.co/SM6MAuiqkB  #tech #news #BYOD #mobile #IT #security</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>941678592970379265</th>\n",
       "      <td>See how @usedgov now investigating @MoskowitzEva &amp;amp; @SuccessCharters for violations of student privacy #FERPA @MarkTreyger718 @Parents4Privacy @FatimaFarax @NYCKidsPAC @jbelluck @CoreyinNYC @Dromm25 @ChalkbeatNY @NYDNBenChapman @katetaylornyt https://t.co/d0PYXySLy9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>289560706506969088</th>\n",
       "      <td>School privacy policies are dangerous. Coworker was not allowed to learn name of student who fractured son's skull due to privacy policy.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1336323095565459461</th>\n",
       "      <td>@susanoha Some edtech is great and others are an abomination, digiLEARN is selling the abomination edtech that is expensive, sunders privacy and harms children. https://t.co/WYSvU3mEM8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1325974559506522112</th>\n",
       "      <td>@PublixHelps The chairs to the right in the grocery store was where 2 people in the pharmacy, first a student and then the pharmacist, tried to give me the flu vaccine. I refused because this is unacceptable. No privacy. Pathetic pt care! https://t.co/t4O0wn4MPf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>854077533104427008</th>\n",
       "      <td>MT @TeriGRight: #CommonCore intrusive data mining threatens student privacy! #KeepYourPromise #StopCommonCore https://t.co/6xyB74ntJa #PJNET</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1101672288985014272</th>\n",
       "      <td>The latest Hecht Hyperborean! https://t.co/YJKNCaAxez Thanks to @catacalypto @twmillar @nevesytrof #privacy #edtech</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>310032618819297281</th>\n",
       "      <td>Allure of \"21st-century technology\" masks privacy-undermining agenda of those allergic to..parental autonomy. http://t.co/gMGlL2xZk4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>310595181432946690</th>\n",
       "      <td>School Official Accused of Accessing Student's Facebook Page #cyber #cyberbullying #privacy... http://t.co/JOb6rYUFOC</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                                                                                                                                                                                                                                              text\n",
       "tweetid                                                                                                                                                                                                                                                                                           \n",
       "101695842209107968                                                                                                                                 Court sides with college accused of snooping on student’s email account http://e74.us/8w #privacy RT @internetcases &lt; Unencrypted = insecure\n",
       "344785106378043393                                                                                                                                                                     BYOD &amp; the implications for the IT team http://t.co/SM6MAuiqkB  #tech #news #BYOD #mobile #IT #security\n",
       "941678592970379265   See how @usedgov now investigating @MoskowitzEva &amp; @SuccessCharters for violations of student privacy #FERPA @MarkTreyger718 @Parents4Privacy @FatimaFarax @NYCKidsPAC @jbelluck @CoreyinNYC @Dromm25 @ChalkbeatNY @NYDNBenChapman @katetaylornyt https://t.co/d0PYXySLy9\n",
       "289560706506969088                                                                                                                                       School privacy policies are dangerous. Coworker was not allowed to learn name of student who fractured son's skull due to privacy policy.\n",
       "1336323095565459461                                                                                       @susanoha Some edtech is great and others are an abomination, digiLEARN is selling the abomination edtech that is expensive, sunders privacy and harms children. https://t.co/WYSvU3mEM8\n",
       "1325974559506522112         @PublixHelps The chairs to the right in the grocery store was where 2 people in the pharmacy, first a student and then the pharmacist, tried to give me the flu vaccine. I refused because this is unacceptable. No privacy. Pathetic pt care! https://t.co/t4O0wn4MPf\n",
       "854077533104427008                                                                                                                                    MT @TeriGRight: #CommonCore intrusive data mining threatens student privacy! #KeepYourPromise #StopCommonCore https://t.co/6xyB74ntJa #PJNET\n",
       "1101672288985014272                                                                                                                                                            The latest Hecht Hyperborean! https://t.co/YJKNCaAxez Thanks to @catacalypto @twmillar @nevesytrof #privacy #edtech\n",
       "310032618819297281                                                                                                                                            Allure of \"21st-century technology\" masks privacy-undermining agenda of those allergic to..parental autonomy. http://t.co/gMGlL2xZk4\n",
       "310595181432946690                                                                                                                                                           School Official Accused of Accessing Student's Facebook Page #cyber #cyberbullying #privacy... http://t.co/JOb6rYUFOC"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df.senti=='negative'].sample(10)[['text']]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "55cb526c-5ce3-4609-af1e-bd04f3dd5226",
   "metadata": {},
   "source": [
    "### Topics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "7c41f7c9-21f9-4695-930b-1426245be8de",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 gauge pseudonym tyranny culprit sarcasm ludicrous nefarious pedophile disingenuous fwiw\n",
      "1 edtech cybersecurity dfir skillsgap nlproc ingramtechsol systemsthinke stcenturyskill thecybersecurityplace cybint\n",
      "2 datum protect sotu schoolprivacyzone paigekowalski datadrivesresult rachelbarrer takectrl replukemesser benjaminbherold\n",
      "3 username share talk story daily event awesome tune david advisor\n",
      "4 school lifeleader schoolchildren safeschool studentvoice chooseprivacy schooldistrict coetail powerschool indyschool\n",
      "5 security byod mobile device enterprisesecurity byodsecurity fiercemobileit securityrisk dellsecurity symantec\n",
      "6 privacy student news washington priva myname privac stud studentright priv\n",
      "7 facebook internship flaw austrian lose harvard expose highlight messenger cancel\n",
      "8 conclude cpdp myriad minefield txcto asugsv sotn bettertogether nuisance rundown\n",
      "9 neighborhood georgetown auburn nazi chattanooga tulsa knife sixth uconn franklin\n"
     ]
    }
   ],
   "source": [
    "priv_topics = [line.split(':')[1] for line in open('priv-tweet-topics.txt')]\n",
    "for pt in range(len(priv_topics)):\n",
    "    print(pt, priv_topics[pt].strip())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "cab7c492-b381-4104-844e-415d58d723d1",
   "metadata": {},
   "outputs": [],
   "source": [
    "interesting_topics=[0, 1,2 ,3,4,6]\n",
    "# 0: talks about tyranny of EdTech\n",
    "# 1: ingramtechsol, thecybersecurityplace, cybint-->general cybersecurity, \n",
    "# 2: names of politicians, journalists who talk about (privacy?) policy\n",
    "# 3: news, articles, announcements, some about how to prevent security/privacy issues \n",
    "# 4: similar to 0\n",
    "# 5: about BOYD, ignore"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "5da494bc-0da7-41be-bad7-97ed4ed83509",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "topic\n",
       " 6    104699\n",
       " 5     37108\n",
       " 1     12735\n",
       " 3      6447\n",
       " 2      1205\n",
       " 4       702\n",
       "-1       699\n",
       " 7        48\n",
       " 0        12\n",
       " 9         3\n",
       "Name: text, dtype: int64"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby('topic').text.count().sort_values(ascending=False)#.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "204d5554-1363-4176-a7f3-5dd053cad6b6",
   "metadata": {},
   "source": [
    "### Sample negative tweets for each topic for manual annotation "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "883b8bdb-933b-4060-bb30-e0dfe4ff9070",
   "metadata": {},
   "outputs": [],
   "source": [
    "# neg_df = df[df.senti=='negative']\n",
    "# neg_df['link']=neg_df.apply(lambda r: 'http://twitter.com/u/status/{}'.format(r.name), axis=1)\n",
    "# neg_df.groupby('topic').text.count().sort_values(ascending=False).head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "1857c8e4-f5ab-4b7e-ac7a-d5e701084038",
   "metadata": {},
   "outputs": [],
   "source": [
    "# neg_df[neg_df.topic<5][['text', 'link']].to_excel('neg-tweet-[0-4].xls')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "1c63795d-d6d9-4ad5-9846-b5b59fe484c8",
   "metadata": {},
   "outputs": [],
   "source": [
    "# tdf= df[df.apply(lambda row: row.topic==5 and 'edtech' in row.text.lower(), axis=1)]\n",
    "# print(len(tdf))\n",
    "# tdf[['text']].to_excel('neg-tweet-5.xls')\n",
    "# tdf.sample(np.min([10,len(tdf)]))[['text', 'topic','topic_cos','created_at']]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "92a064e7-1c06-4d7c-988f-c84632e5a0d2",
   "metadata": {},
   "source": [
    "### remove duplicate tweets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 204,
   "id": "d84faa10-856e-4ea0-9ac0-5c580c9748e6",
   "metadata": {},
   "outputs": [],
   "source": [
    "# topic_6 = pd.read_excel('neg-tweet-6.xlsx')#pd.concat([pd.read_excel('neg-tweet-6.xlsx').iloc[:1753], pd.read_excel('neg-tweet-6-no-dup.xlsx')[1753:]])\n",
    "# print(topic_6.shape)\n",
    "# dups = [line.strip() for line in open('duplicates.txt')][0].split(',')\n",
    "# dups=list(filter(None, dups))\n",
    "# dup_tweets = []\n",
    "# for dup in dups:\n",
    "#     dup_tweets += list(topic_6[topic_6.apply(lambda row: dup.strip().lower() in row.text.lower() and \\\n",
    "#                                              (isinstance(row.Code1, float) or len(row.Code1)==0), axis=1)].index)\n",
    "# len(dup_tweets)\n",
    "# topic_6[topic_6.apply(lambda row: ' rfid ' in row.text.lower(), axis=1)] "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "9d1be493-2a2e-4fa5-8f3f-d0208b5558b9",
   "metadata": {},
   "outputs": [],
   "source": [
    "# no_dup=topic_6.drop(index=dup_tweets)\n",
    "# print(no_dup.shape)\n",
    "# no_dup[['text','Code1', 'link']].to_excel('neg-tweet-6-final.xlsx')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3b9c5936-6c92-4d6b-95bb-a5285f7d45a8",
   "metadata": {},
   "source": [
    "### Analyze labelled tweets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "85a3db46-b0f6-452f-a3f6-dd3b185614a5",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_tfidf_top_features(documents,n_top=10, max_df=.3, min_df=5):\n",
    "    tfidf_vectorizer = TfidfVectorizer(max_df=max_df, min_df=min_df,stop_words='english')\n",
    "    tfidf = tfidf_vectorizer.fit_transform(documents)\n",
    "    importance = np.argsort(np.asarray(tfidf.sum(axis=0)).ravel())[::-1]\n",
    "    tfidf_feature_names = np.array(tfidf_vectorizer.get_feature_names())\n",
    "    return tfidf_feature_names[importance[:n_top]]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "3bc6c29a-33a0-4162-9870-c8ddf2e5de57",
   "metadata": {},
   "outputs": [],
   "source": [
    "# labled_df = pd.concat([pd.read_csv('neg-tweet-6-final.csv',  sep=',', encoding = \"utf-8\"),\n",
    "#                       pd.read_csv('neg-tweet-(0-4).csv',  sep=',', encoding = \"utf-8\")])\n",
    "# labled_df['tweetid'] = labled_df.apply(lambda row: row.link.split('/')[-1], axis=1)\n",
    "# labled_df = labled_df[labled_df.apply(lambda row: isinstance(row.Code, str) and len(row.Code.strip())>2, axis=1)]\n",
    "\n",
    "# df= pd.read_csv('neg-tweet-5.csv',  sep=',', encoding = \"utf-8\")\n",
    "# df['tweetid']=df.tweetid.astype(str)\n",
    "# labled_df=pd.concat([labled_df, df])\n",
    "\n",
    "# labled_df.set_index('tweetid', inplace=True)\n",
    "# print(labled_df.shape)\n",
    "# labled_df=labled_df[['text', 'Code']]\n",
    "# labled_df.to_csv('all-sec-priv-labeled-tweets.csv', encoding='utf-8')\n",
    "# labled_df.tweetid=labled_df.tweetid.astype(str)\n",
    "# labled_df.set_index('tweetid', inplace=True)\n",
    "\n",
    "# ldf=df.loc[labled_df.index]\n",
    "# ldf['code'] = labled_df.code\n",
    "# ldf.shape\n",
    "# ldf.to_pickle('all-sec-priv-labeled-tweets.pkl')\n",
    "# ldf.to_csv('all-sec-priv-labeled-tweets.csv', encoding='utf-8')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "d683876e-624b-41f0-aa90-64d03b8558fc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(4648, 34)\n"
     ]
    },
    {
     "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>author_id</th>\n",
       "      <th>text</th>\n",
       "      <th>created_at</th>\n",
       "      <th>geo</th>\n",
       "      <th>retweets</th>\n",
       "      <th>replies</th>\n",
       "      <th>likes</th>\n",
       "      <th>quote_count</th>\n",
       "      <th>lang</th>\n",
       "      <th>conversation_id</th>\n",
       "      <th>...</th>\n",
       "      <th>author_group</th>\n",
       "      <th>search_grp</th>\n",
       "      <th>entities_parsed</th>\n",
       "      <th>urls</th>\n",
       "      <th>hashtags</th>\n",
       "      <th>anno_type</th>\n",
       "      <th>anno_text</th>\n",
       "      <th>mentions</th>\n",
       "      <th>tweet_type</th>\n",
       "      <th>code</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>tweetid</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1301656105626476546</th>\n",
       "      <td>1296874226100494338</td>\n",
       "      <td>I’m a teacher. If I mishandled one student, one time, I’d no longer have a job &amp;amp; no longer be able to work w kids. &amp;amp; Yet cops kill &amp;amp; abuse over &amp;amp; over again, while keeping their jobs, their privacy, &amp;amp; their pay. \\nIt amazes me that #defundthepolice is controversial or political</td>\n",
       "      <td>2020-09-03 22:59:33+00:00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>en</td>\n",
       "      <td>1301656105626476546</td>\n",
       "      <td>...</td>\n",
       "      <td>unknown</td>\n",
       "      <td>other</td>\n",
       "      <td>{}</td>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>novel</td>\n",
       "      <td>other</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1312039409689255936</th>\n",
       "      <td>705983975957745664</td>\n",
       "      <td>‘It Feels Like an Invasion of Privacy’ ➡️ THAT'S BECAUSE IT IS. \\n\\nIf one's assessment materials (i.e., an exam) are easy enough to game that you can catch cheating by surveillance software, you're not educating in the first place.\\n\\nhttps://t.co/IYWN5qeFaO</td>\n",
       "      <td>2020-10-02 14:39:06+00:00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>en</td>\n",
       "      <td>1312039409689255936</td>\n",
       "      <td>...</td>\n",
       "      <td>academic</td>\n",
       "      <td>other</td>\n",
       "      <td>{'urls': [{'start': 232, 'end': 255, 'url': 'https://t.co/IYWN5qeFaO', 'expanded_url': 'https://www.nytimes.com/2020/09/29/style/testing-schools-proctorio.html', 'display_url': 'nytimes.com/2020/09/29/sty…'}]}</td>\n",
       "      <td>[www.nytimes.com]</td>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>novel</td>\n",
       "      <td>privacy-generic</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2 rows × 34 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                               author_id  \\\n",
       "tweetid                                    \n",
       "1301656105626476546  1296874226100494338   \n",
       "1312039409689255936   705983975957745664   \n",
       "\n",
       "                                                                                                                                                                                                                                                                                                                           text  \\\n",
       "tweetid                                                                                                                                                                                                                                                                                                                           \n",
       "1301656105626476546  I’m a teacher. If I mishandled one student, one time, I’d no longer have a job &amp; no longer be able to work w kids. &amp; Yet cops kill &amp; abuse over &amp; over again, while keeping their jobs, their privacy, &amp; their pay. \\nIt amazes me that #defundthepolice is controversial or political   \n",
       "1312039409689255936                                         ‘It Feels Like an Invasion of Privacy’ ➡️ THAT'S BECAUSE IT IS. \\n\\nIf one's assessment materials (i.e., an exam) are easy enough to game that you can catch cheating by surveillance software, you're not educating in the first place.\\n\\nhttps://t.co/IYWN5qeFaO   \n",
       "\n",
       "                                   created_at  geo  retweets  replies  likes  \\\n",
       "tweetid                                                                        \n",
       "1301656105626476546 2020-09-03 22:59:33+00:00  NaN         0        0      0   \n",
       "1312039409689255936 2020-10-02 14:39:06+00:00  NaN         0        0      0   \n",
       "\n",
       "                     quote_count lang      conversation_id  ... author_group  \\\n",
       "tweetid                                                     ...                \n",
       "1301656105626476546            0   en  1301656105626476546  ...      unknown   \n",
       "1312039409689255936            0   en  1312039409689255936  ...     academic   \n",
       "\n",
       "                    search_grp  \\\n",
       "tweetid                          \n",
       "1301656105626476546      other   \n",
       "1312039409689255936      other   \n",
       "\n",
       "                                                                                                                                                                                                                       entities_parsed  \\\n",
       "tweetid                                                                                                                                                                                                                                  \n",
       "1301656105626476546                                                                                                                                                                                                                 {}   \n",
       "1312039409689255936  {'urls': [{'start': 232, 'end': 255, 'url': 'https://t.co/IYWN5qeFaO', 'expanded_url': 'https://www.nytimes.com/2020/09/29/style/testing-schools-proctorio.html', 'display_url': 'nytimes.com/2020/09/29/sty…'}]}   \n",
       "\n",
       "                                  urls hashtags anno_type anno_text mentions  \\\n",
       "tweetid                                                                        \n",
       "1301656105626476546                 []       []        []        []       []   \n",
       "1312039409689255936  [www.nytimes.com]       []        []        []       []   \n",
       "\n",
       "                    tweet_type             code  \n",
       "tweetid                                          \n",
       "1301656105626476546      novel            other  \n",
       "1312039409689255936      novel  privacy-generic  \n",
       "\n",
       "[2 rows x 34 columns]"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "labled_df = pd.read_pickle('all-sec-priv-labeled-tweets.pkl')\n",
    "print(labled_df.shape)\n",
    "labled_df.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "5a9e893c-3a37-467c-9195-2547e54b79d5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "code\n",
       "other-student-privacy    864\n",
       "privacy-specific         725\n",
       "other-privacy            574\n",
       "article                  494\n",
       "other                    424\n",
       "privacy-generic          406\n",
       "policy                   262\n",
       "privacy incident         190\n",
       "info_brand               131\n",
       "tech_brandi               86\n",
       "edtech-generic            76\n",
       "edtech-ques               45\n",
       "edtech ban                 4\n",
       "Name: text, dtype: int64"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "labled_df.groupby('code').text.count().sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "id": "df942f47-916e-469e-8585-f63cb44456f8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x1bf56f4f0>"
      ]
     },
     "execution_count": 144,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for code in ['privacy-generic', 'privacy-specific', 'privacy incident', 'policy', 'other-student-privacy', 'other-privacy', 'article']:\n",
    "    labled_df[labled_df.code==code].reset_index().set_index('created_at').groupby(pd.Grouper(freq='3M')).text.count().plot(label=code, linewidth=3, figsize=(16,8))\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "b5b96c3f-66a0-4cf6-a26f-d18c9b0a9133",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "code                   author_group   \n",
       "article                personal           6\n",
       "                       techie             5\n",
       "edtech-ques            academic           5\n",
       "privacy incident       personal           5\n",
       "edtech-ques            business_brand     5\n",
       "other-student-privacy  techie             4\n",
       "tech_brandi            news               4\n",
       "                       academic           3\n",
       "                       techie             3\n",
       "edtech ban             academic_techie    2\n",
       "privacy incident       techie             2\n",
       "edtech-generic         personal           2\n",
       "edtech-ques            personal_brand     2\n",
       "info_brand             academic           2\n",
       "                       techie             2\n",
       "privacy-generic        techie             1\n",
       "policy                 techie             1\n",
       "edtech-ques            news               1\n",
       "                       personal           1\n",
       "info_brand             news               1\n",
       "Name: text, dtype: int64"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "labled_df[labled_df.author_group!='unknown'].groupby(['code', 'author_group']).text.count().sort_values(ascending=False).tail(20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "1f252066-6d41-41c1-8d3b-7fa161df4a47",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\\begin{tabular}{llr}\n",
      "\\toprule\n",
      "                 &        &  text \\\\\n",
      "code & author\\_group &       \\\\\n",
      "\\midrule\n",
      "privacy-generic & academic &    33 \\\\\n",
      "                 & academic\\_techie &    86 \\\\\n",
      "                 & business\\_brand &    29 \\\\\n",
      "                 & news &     7 \\\\\n",
      "                 & personal &     6 \\\\\n",
      "                 & personal\\_brand &    25 \\\\\n",
      "                 & techie &     1 \\\\\n",
      "privacy-specific & academic &    80 \\\\\n",
      "                 & academic\\_techie &   125 \\\\\n",
      "                 & business\\_brand &    44 \\\\\n",
      "                 & news &    16 \\\\\n",
      "                 & personal &    13 \\\\\n",
      "                 & personal\\_brand &    34 \\\\\n",
      "                 & techie &    21 \\\\\n",
      "\\bottomrule\n",
      "\\end{tabular}\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/1p/0rwj4hf522b5v3bzw33bsdqw0000gn/T/ipykernel_86551/2715073796.py:2: 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(labled_df[(labled_df.author_group!='unknown') &((labled_df.code=='privacy-generic') | (labled_df.code=='privacy-specific'))].groupby(['code', 'author_group']).text.count().to_frame().to_latex())\n"
     ]
    }
   ],
   "source": [
    "'''who tweeted generic and specific privacy concerns'''\n",
    "print(labled_df[(labled_df.author_group!='unknown') &((labled_df.code=='privacy-generic') | (labled_df.code=='privacy-specific'))].groupby(['code', 'author_group']).text.count().to_frame().to_latex())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "c07b91f7-24df-4a5b-8af2-35a6df641f14",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "privacy-generic,academic:  username, privacy, student, edtech, surveillance, datum, law, time, platform, issue\n",
      "privacy-generic,academic_techie:  username, privacy, student, edtech, datum, school, surveillance, issue, app, teacher\n",
      "privacy-generic,business_brand:  privacy, username, student, datum, school, edtech, byod, surveillance, onlinesafety, security\n",
      "privacy-generic,personal_brand:  privacy, username, student, datum, edtech, eddata, teacher, concern, school, issue\n",
      "privacy-generic,personal:  username, student, expectation, accountability, understand, secure, threaten, pjnet, keepyourpromise, intrusive\n",
      "privacy-generic,news:  student, mooc, loss, major, drawback, blow, surveillance, username, people, rep\n",
      "\n",
      "privacy-specific,academic:  username, student, privacy, issue, datum, surveillance, concern, school, violation, teacher\n",
      "privacy-specific,academic_techie:  username, privacy, student, edtech, school, issue, security, google, surveillance, app\n",
      "privacy-specific,business_brand:  privacy, username, student, school, facebook, issue, security, concern, kid, reliance\n",
      "privacy-specific,personal_brand:  username, privacy, student, edtech, datum, surveillance, breach, security, school, issue\n",
      "privacy-specific,personal:  username, student, violate, edtech, zoom, time, assumption, guilt, right, danger\n",
      "privacy-specific,news:  student, privacy, surveillance, issue, online, username, exam, unfair, global, spark\n",
      "privacy-specific,techie:  username, security, privacy, exam, student, delete, business, fail, information, account\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/admin/work/edtech-scrape/edenv/lib/python3.9/site-packages/sklearn/utils/deprecation.py:87: FutureWarning: Function get_feature_names is deprecated; get_feature_names is deprecated in 1.0 and will be removed in 1.2. Please use get_feature_names_out instead.\n",
      "  warnings.warn(msg, category=FutureWarning)\n"
     ]
    }
   ],
   "source": [
    "d=dict()\n",
    "common=set()\n",
    "for t in ['privacy-generic', 'privacy-specific']:\n",
    "    for a in ['academic', 'academic_techie', 'business_brand', 'personal_brand', 'personal','news','techie']:\n",
    "        text = labled_df[(labled_df.author_group==a) & (labled_df.code==t)].text_clean_topic.values\n",
    "        if len(text) > 1:\n",
    "            tfidf = get_tfidf_top_features(text, n_top=10, max_df=.8, min_df=1)\n",
    "            print('{}:  {}'.format(t+','+a, ', '.join(tfidf)))\n",
    "    print()\n",
    "            # d[t+','+a]= tfidf\n",
    "            # if len(common)==0:\n",
    "            #     common.update(set(tfidf))\n",
    "            # else:\n",
    "            #     common=common.intersection(tfidf)\n",
    "# for k in d.keys():\n",
    "#     print('{}:  {}'.format(k, ', '.join(d[k])))\n",
    "    #print('\\t\\t{}'.format(set(d[k]).difference(common)))\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f546265e-7148-4406-af8c-ed837bc95cd1",
   "metadata": {},
   "source": [
    "### Visibility"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "08c4cc61-a31b-41c4-86e1-96f00740337c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "unique authors:  3577\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "count    3577.000000\n",
       "mean        1.299413\n",
       "std         2.248435\n",
       "min         1.000000\n",
       "25%         1.000000\n",
       "50%         1.000000\n",
       "75%         1.000000\n",
       "max        76.000000\n",
       "Name: text, dtype: float64"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'''Were a few people overly representitive?'''\n",
    "print('unique authors: ', len(labled_df.author_id.unique()))\n",
    "labled_df.groupby(['author_id']).text.count().describe()#sort_values(ascending=False).head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "0f6a1461-971c-4878-aa1a-aa4d344018e3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "author_id\n",
       "36991951              76\n",
       "118530377             59\n",
       "19895837              45\n",
       "224059040             37\n",
       "2968343710            33\n",
       "216928966             24\n",
       "2299828074            23\n",
       "90749008              22\n",
       "15694680              17\n",
       "182539117             17\n",
       "2533386684            16\n",
       "768098863395827712    15\n",
       "6505892               15\n",
       "739293                11\n",
       "3959010562            11\n",
       "7044082               11\n",
       "26744205              11\n",
       "16946286              11\n",
       "32866389              11\n",
       "2586135432            10\n",
       "4256298574            10\n",
       "15740610              10\n",
       "Name: text, dtype: int64"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "labled_df.groupby(['author_id']).text.count().sort_values(ascending=False).head(22)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "9b813657-126a-4d22-9fd4-0fd45d4bdf44",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "'''Retweets and likes on tweets expressing generic and specific concerns'''\n",
    "plt.figure(figsize=(16, 8))\n",
    "markers={'privacy-generic':'D', 'privacy-specific': 'o', 'privacy incident':'*'}\n",
    "g=labled_df[labled_df.code==code].reset_index().set_index('created_at').groupby(pd.Grouper(freq='6M'))\n",
    "for code in ['privacy-generic', 'privacy-specific', 'privacy incident']:\n",
    "    g = labled_df[labled_df.code==code].reset_index().set_index('created_at').groupby(pd.Grouper(freq='3M'))#.text.count().plot(label=code, marker=\"o\", markersize=15, linestyle='none')\n",
    "\n",
    "    gdf = g.text.agg(['count'])\n",
    "    gdf['retweets'] = g.retweets.mean()\n",
    "    gdf['likes'] = g.likes.mean()\n",
    "    gdf.reset_index(inplace=True)\n",
    "    sns.scatterplot(\n",
    "        data=gdf, x='created_at', y='count',  size=\"retweets\", label=code, marker='o',#markers[code],\n",
    "        sizes=(30, 300), hue_norm=(0, 7)#, legend=\"full\"\n",
    "    )\n",
    "\n",
    "plt.legend(title=\"\", title_fontsize=12, fontsize=11, fancybox=True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "id": "5dd424f0-02ef-497e-93a8-7579d3d2c641",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "code\n",
       "edtech ban               232.750000\n",
       "privacy-specific           3.914483\n",
       "privacy-generic            1.820197\n",
       "other-student-privacy      1.500000\n",
       "other-privacy              1.465157\n",
       "article                    1.338057\n",
       "info_brand                 1.244275\n",
       "other                      0.981132\n",
       "privacy incident           0.978947\n",
       "policy                     0.977099\n",
       "tech_brandi                0.860465\n",
       "edtech-generic             0.578947\n",
       "edtech-ques                0.400000\n",
       "Name: retweets, dtype: float64"
      ]
     },
     "execution_count": 108,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "labled_df.groupby('code').retweets.mean().sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "id": "b523c6f2-80d2-4b24-9893-09d492a0a145",
   "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>text</th>\n",
       "      <th>retweets</th>\n",
       "      <th>likes</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>tweetid</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1363966363602051074</th>\n",
       "      <td>So I've been stuck on the camera policy for my Zoom classes. I totally get the worries about student privacy and invasiveness. But I also know that my teaching gets radically worse when I'm facing mostly all black squares. So this time, I tried complete honesty.\\n\\nReport-thread:</td>\n",
       "      <td>930</td>\n",
       "      <td>4782</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1207861674264412160</th>\n",
       "      <td>@33heupel Depending on what you can use in regards to Ss data privacy @Flipgrid could be a great tool here. #ditchbook</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1249694208471863297</th>\n",
       "      <td>@Garreth_Wigg @GoogleForEdu @MicrosoftEDU @AppleEDU yes they do have the same problem, a lot of these systems and services are restricted in our public schools. In the private sector however it is not so restrictive. new data privacy laws have had wide sweeping implications on how we obtain perm...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1357788885015789570</th>\n",
       "      <td>So sad (and a bit surprised) that we are not able to suggest that our teachers use @Flipgrid anymore, since they won't sign our state student data privacy agreement. We love Flipgrid!💚</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                                                                                                                                                                                                                                                                            text  \\\n",
       "tweetid                                                                                                                                                                                                                                                                                                                            \n",
       "1363966363602051074                     So I've been stuck on the camera policy for my Zoom classes. I totally get the worries about student privacy and invasiveness. But I also know that my teaching gets radically worse when I'm facing mostly all black squares. So this time, I tried complete honesty.\\n\\nReport-thread:   \n",
       "1207861674264412160                                                                                                                                                                                       @33heupel Depending on what you can use in regards to Ss data privacy @Flipgrid could be a great tool here. #ditchbook   \n",
       "1249694208471863297  @Garreth_Wigg @GoogleForEdu @MicrosoftEDU @AppleEDU yes they do have the same problem, a lot of these systems and services are restricted in our public schools. In the private sector however it is not so restrictive. new data privacy laws have had wide sweeping implications on how we obtain perm...   \n",
       "1357788885015789570                                                                                                                     So sad (and a bit surprised) that we are not able to suggest that our teachers use @Flipgrid anymore, since they won't sign our state student data privacy agreement. We love Flipgrid!💚   \n",
       "\n",
       "                     retweets  likes  \n",
       "tweetid                               \n",
       "1363966363602051074       930   4782  \n",
       "1207861674264412160         0      4  \n",
       "1249694208471863297         0      0  \n",
       "1357788885015789570         1      4  "
      ]
     },
     "execution_count": 102,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "labled_df[labled_df.code=='edtech ban'][['text', 'retweets', 'likes']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "id": "54e42e5e-1d57-4445-be67-cc38f744da99",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "code\n",
       "edtech ban               1197.500000\n",
       "privacy-specific            9.673103\n",
       "privacy-generic             5.529557\n",
       "other-student-privacy       5.171296\n",
       "other                       4.577830\n",
       "other-privacy               3.869338\n",
       "edtech-generic              2.157895\n",
       "policy                      1.801527\n",
       "privacy incident            1.710526\n",
       "info_brand                  1.618321\n",
       "article                     1.566802\n",
       "tech_brandi                 1.244186\n",
       "edtech-ques                 1.222222\n",
       "Name: likes, dtype: float64"
      ]
     },
     "execution_count": 103,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "labled_df.groupby('code').likes.mean().sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "id": "8c44aa81-130b-46b4-a245-0604b43fc6c4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "author_id\n",
       "36991951      76\n",
       "118530377     59\n",
       "19895837      45\n",
       "224059040     37\n",
       "2968343710    33\n",
       "216928966     24\n",
       "2299828074    23\n",
       "90749008      22\n",
       "15694680      17\n",
       "182539117     17\n",
       "Name: text, dtype: int64"
      ]
     },
     "execution_count": 102,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "labled_df.groupby('author_id').text.count().sort_values(ascending=False).head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "458bd930-6a1b-45ef-b968-d664ac1e3d2a",
   "metadata": {},
   "source": [
    "### Keywords in each group"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "1fee3c12-2945-4147-8245-1ec779f0edda",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['username', 'privacy', 'student', 'edtech', 'datum',\n",
       "       'surveillance', 'school', 'issue', 'eddata', 'teacher', 'concern',\n",
       "       'app', 'kid', 'security', 'parent', 'education', 'bad', 'protect',\n",
       "       'child', 'online', 'tool', 'company', 'time', 'learn',\n",
       "       'information'], dtype='<U17')"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "550b381e-7702-4de7-889e-b293f7a190ae",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['username', 'privacy', 'student', 'school', 'surveillance',\n",
       "       'edtech', 'security', 'datum', 'issue', 'software', 'invasion',\n",
       "       'app', 'violation', 'google', 'concern', 'exam', 'policy', 'kid',\n",
       "       'zoom', 'teacher', 'bad', 'online', 'violate', 'proctorio', 'feel'],\n",
       "      dtype='<U20')"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "db366772-0dec-4253-978f-decf0e4290f6",
   "metadata": {},
   "source": [
    "### Entities in each group"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 187,
   "id": "14074f4e-43d8-40d6-b6cb-b0119b6b9d4a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "unique mentions: 129\n",
      "[('goguardian', 8), ('chegg', 7), ('examsoft', 6), ('edpuzzle', 5), ('pearsonvue', 4), ('linkletter', 3), ('googleforedu', 3), ('coursehero', 3), ('chegghelp', 3), ('classdojo', 2)]\n",
      "unique mentions: 32\n",
      "[('zdnet', 1), ('danielharvey9', 1), ('kmp39', 1), ('jeremycyoung', 1), ('edmodo', 1), ('mikemcsharry', 1), ('eileenbelastock', 1), ('edtechdigest', 1), ('douglevin', 1), ('managedmethods', 1)]\n"
     ]
    }
   ],
   "source": [
    "'''what edtech providers were mentioned most in privacy groups'''\n",
    "def common_mentions(kw):\n",
    "    mentions = labled_df[labled_df.code==kw].mentions\n",
    "    mentions = list(itertools.chain.from_iterable(mentions))\n",
    "    mentions = [m.lower() for m in mentions]\n",
    "    print(\"unique mentions:\", len(set(mentions)))\n",
    "    mentions = Counter(mentions)\n",
    "    print(mentions.most_common(10))\n",
    "common_mentions('privacy-specific')\n",
    "common_mentions('privacy-generic')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 175,
   "id": "675bc96c-aff1-4786-8fe5-9c40cc655c3d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tweetid\n",
       "1339974630698856448    []\n",
       "199919127895351296     []\n",
       "540992121952423936     []\n",
       "672312661560848384     []\n",
       "561143099498721280     []\n",
       "                       ..\n",
       "183562595104600066     []\n",
       "200969939442610177     []\n",
       "342299881228099584     []\n",
       "605423600452124674     []\n",
       "512007017008218112     []\n",
       "Name: urls, Length: 494, dtype: object"
      ]
     },
     "execution_count": 175,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'''what urls/domains are used for article group'''\n",
    "urls = labled_df[labled_df.code=='article'].urls\n",
    "urls"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 195,
   "id": "f6a2b516-d351-403c-98b4-c9a2b1a1521e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tweetid\n",
       "1301656105626476546    negative\n",
       "1312039409689255936    negative\n",
       "570992748725194752     negative\n",
       "1006594055432802305    negative\n",
       "1352870101775654913    negative\n",
       "                         ...   \n",
       "245581393768366080     positive\n",
       "245338142209605632      neutral\n",
       "245238058423828480      neutral\n",
       "245212511979319296      neutral\n",
       "1024331454820446209     neutral\n",
       "Name: senti, Length: 4648, dtype: object"
      ]
     },
     "execution_count": 195,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "labled_df.senti"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "bf0ae1cc-a278-424c-9834-8f661e072dbd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "56"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x='@Schoology, @Schoolytics, @GoogleForEdu, @moodle, @MyPowerSchool, @CanvasLMS, @Educ_Technology, @EdTech_K12, @MerlynMind, @edXOnline, @DigitalPromise, @LearningAtScale, @edtechdigest, @edmodo, @ClassDojo, @Newsela, @quizizz, @blackboard, @edpuzzle, @PearDeck, @Flipgrid, @BookCreatorApp, @mergeedu, @BookCreatorApp, @Seesaw, @TeachFX, @udacity, @coursera, @udemy, @edXOnline, @MicrosoftEDU, @proctorio, @examity, @ProctortrackPro, @speedexam, @in_examonline, @MercerMettl, @Examus1, @ProctorU, @Verificient, @AIProctor, @ExamSoft, @conductexam, @ProctorExam, @PSIServicesLLC, @pearsonvue, @merittracsvc, @Civitaslearning, @DreamBox_Learn, @GoGuardian, @GuildEducation, @Instructure, @Chegg, @CreativeLive, @CourseHero, @Masterclass'\n",
    "len(x.split(','))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ae28ceec-e54b-44d1-9a7f-f785f9af5a05",
   "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": "3.9.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}