03_text_analysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.feature_extraction.text import CountVectorizer, TfidfVectorizer\n",
    "import seaborn as sns\n",
    "from matplotlib import pyplot as plt\n",
    "from matplotlib import style\n",
    "from wordcloud import WordCloud\n",
    "import pyLDAvis, pyLDAvis.sklearn\n",
    "from sklearn.manifold import TSNE\n",
    "from sklearn.cluster import KMeans\n",
    "from sklearn.decomposition import LatentDirichletAllocation, PCA\n",
    "pyLDAvis.enable_notebook()\n",
    "style.use('fivethirtyeight')\n",
    "import re\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>title</th>\n",
       "      <th>date</th>\n",
       "      <th>text</th>\n",
       "      <th>clean_text</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>Talk to the People of President Rodrigo Roa Du...</td>\n",
       "      <td>2020-05-04</td>\n",
       "      <td>\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...</td>\n",
       "      <td>president duterte reduced changed way life eve...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>Talk to the People of President Rodrigo Roa Du...</td>\n",
       "      <td>2020-04-27</td>\n",
       "      <td>\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...</td>\n",
       "      <td>dost secretary fortunato dela pea susunod kahi...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>Talk to the People of President Rodrigo Roa Du...</td>\n",
       "      <td>2020-04-24</td>\n",
       "      <td>\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...</td>\n",
       "      <td>presidential spokesperson harry roque magandan...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>Talk to the People of President Rodrigo Roa Du...</td>\n",
       "      <td>2020-04-16</td>\n",
       "      <td>\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...</td>\n",
       "      <td>mga kapatid kong pilipino nandito harap ninyo ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>Intervention of President Rodrigo Roa Duterte ...</td>\n",
       "      <td>2020-04-14</td>\n",
       "      <td>\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...</td>\n",
       "      <td>thank mr chair majesty excellencies welcome le...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                               title        date  \\\n",
       "0  Talk to the People of President Rodrigo Roa Du...  2020-05-04   \n",
       "1  Talk to the People of President Rodrigo Roa Du...  2020-04-27   \n",
       "2  Talk to the People of President Rodrigo Roa Du...  2020-04-24   \n",
       "3  Talk to the People of President Rodrigo Roa Du...  2020-04-16   \n",
       "4  Intervention of President Rodrigo Roa Duterte ...  2020-04-14   \n",
       "\n",
       "                                                text  \\\n",
       "0  \\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...   \n",
       "1  \\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...   \n",
       "2  \\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...   \n",
       "3  \\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...   \n",
       "4  \\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...   \n",
       "\n",
       "                                          clean_text  \n",
       "0  president duterte reduced changed way life eve...  \n",
       "1  dost secretary fortunato dela pea susunod kahi...  \n",
       "2  presidential spokesperson harry roque magandan...  \n",
       "3  mga kapatid kong pilipino nandito harap ninyo ...  \n",
       "4  thank mr chair majesty excellencies welcome le...  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv('data/tika_transcript_clean.csv', index_col=0)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Plotting Speech Frequency"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>title</th>\n",
       "      <th>date</th>\n",
       "      <th>text</th>\n",
       "      <th>clean_text</th>\n",
       "      <th>date_diff</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>Speech of President Rodrigo Roa Duterte during...</td>\n",
       "      <td>2018-03-18</td>\n",
       "      <td>\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...</td>\n",
       "      <td>kindly sit thank courtesy troop commander coul...</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>Speech of President Rodrigo Roa Duterte during...</td>\n",
       "      <td>2018-03-20</td>\n",
       "      <td>\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...</td>\n",
       "      <td>kindly sit salamat ho kortesiya ninyo troop co...</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>Speech of President Rodrigo Roa Duterte during...</td>\n",
       "      <td>2018-03-20</td>\n",
       "      <td>\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...</td>\n",
       "      <td>kindly sit thank courtesy ladies gentlemen usu...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>Speech of President Rodrigo Roa Duterte during...</td>\n",
       "      <td>2018-03-21</td>\n",
       "      <td>\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...</td>\n",
       "      <td>kindly sit thank courtesy troop commander coul...</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>Speech of President Rodrigo Roa Duterte during...</td>\n",
       "      <td>2018-03-21</td>\n",
       "      <td>\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...</td>\n",
       "      <td>duterte duterte hinanap kayo noong election ni...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>Speech of President Rodrigo Roa Duterte during...</td>\n",
       "      <td>2018-03-23</td>\n",
       "      <td>\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...</td>\n",
       "      <td>salamat panglingkod mo kanang una tanan kinaha...</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>Speech Of President Rodrigo Roa Duterte during...</td>\n",
       "      <td>2018-03-26</td>\n",
       "      <td>\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...</td>\n",
       "      <td>thank courtesy kindly sit prepared speech two ...</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>Speech of President Rodrigo Roa Duterte during...</td>\n",
       "      <td>2018-04-02</td>\n",
       "      <td>\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...</td>\n",
       "      <td>kindly sit salamat courtesy ninyo [bakit dumid...</td>\n",
       "      <td>7.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                               title       date  \\\n",
       "0  Speech of President Rodrigo Roa Duterte during... 2018-03-18   \n",
       "1  Speech of President Rodrigo Roa Duterte during... 2018-03-20   \n",
       "2  Speech of President Rodrigo Roa Duterte during... 2018-03-20   \n",
       "3  Speech of President Rodrigo Roa Duterte during... 2018-03-21   \n",
       "4  Speech of President Rodrigo Roa Duterte during... 2018-03-21   \n",
       "5  Speech of President Rodrigo Roa Duterte during... 2018-03-23   \n",
       "6  Speech Of President Rodrigo Roa Duterte during... 2018-03-26   \n",
       "7  Speech of President Rodrigo Roa Duterte during... 2018-04-02   \n",
       "\n",
       "                                                text  \\\n",
       "0  \\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...   \n",
       "1  \\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...   \n",
       "2  \\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...   \n",
       "3  \\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...   \n",
       "4  \\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...   \n",
       "5  \\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...   \n",
       "6  \\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...   \n",
       "7  \\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...   \n",
       "\n",
       "                                          clean_text  date_diff  \n",
       "0  kindly sit thank courtesy troop commander coul...        NaN  \n",
       "1  kindly sit salamat ho kortesiya ninyo troop co...        2.0  \n",
       "2  kindly sit thank courtesy ladies gentlemen usu...        0.0  \n",
       "3  kindly sit thank courtesy troop commander coul...        1.0  \n",
       "4  duterte duterte hinanap kayo noong election ni...        0.0  \n",
       "5  salamat panglingkod mo kanang una tanan kinaha...        2.0  \n",
       "6  thank courtesy kindly sit prepared speech two ...        3.0  \n",
       "7  kindly sit salamat courtesy ninyo [bakit dumid...        7.0  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = df.sort_values('date').reset_index().drop('index', axis=1)\n",
    "df['date'] = pd.to_datetime(df['date'], format='%Y-%m-%d')\n",
    "df['date_diff'] = df['date'].diff().dt.days\n",
    "\n",
    "df.head(8)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>title</th>\n",
       "      <th>date</th>\n",
       "      <th>text</th>\n",
       "      <th>clean_text</th>\n",
       "      <th>date_diff</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>414</td>\n",
       "      <td>Intervention of President Rodrigo Roa Duterte ...</td>\n",
       "      <td>2020-04-14</td>\n",
       "      <td>\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...</td>\n",
       "      <td>thank mr chair majesty excellencies welcome le...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>415</td>\n",
       "      <td>Talk to the People of President Rodrigo Roa Du...</td>\n",
       "      <td>2020-04-16</td>\n",
       "      <td>\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...</td>\n",
       "      <td>mga kapatid kong pilipino nandito harap ninyo ...</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>416</td>\n",
       "      <td>Talk to the People of President Rodrigo Roa Du...</td>\n",
       "      <td>2020-04-24</td>\n",
       "      <td>\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...</td>\n",
       "      <td>presidential spokesperson harry roque magandan...</td>\n",
       "      <td>8.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>417</td>\n",
       "      <td>Talk to the People of President Rodrigo Roa Du...</td>\n",
       "      <td>2020-04-27</td>\n",
       "      <td>\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...</td>\n",
       "      <td>dost secretary fortunato dela pea susunod kahi...</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>418</td>\n",
       "      <td>Talk to the People of President Rodrigo Roa Du...</td>\n",
       "      <td>2020-05-04</td>\n",
       "      <td>\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...</td>\n",
       "      <td>president duterte reduced changed way life eve...</td>\n",
       "      <td>7.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 title       date  \\\n",
       "414  Intervention of President Rodrigo Roa Duterte ... 2020-04-14   \n",
       "415  Talk to the People of President Rodrigo Roa Du... 2020-04-16   \n",
       "416  Talk to the People of President Rodrigo Roa Du... 2020-04-24   \n",
       "417  Talk to the People of President Rodrigo Roa Du... 2020-04-27   \n",
       "418  Talk to the People of President Rodrigo Roa Du... 2020-05-04   \n",
       "\n",
       "                                                  text  \\\n",
       "414  \\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...   \n",
       "415  \\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...   \n",
       "416  \\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...   \n",
       "417  \\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...   \n",
       "418  \\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...   \n",
       "\n",
       "                                            clean_text  date_diff  \n",
       "414  thank mr chair majesty excellencies welcome le...        0.0  \n",
       "415  mga kapatid kong pilipino nandito harap ninyo ...        2.0  \n",
       "416  presidential spokesperson harry roque magandan...        8.0  \n",
       "417  dost secretary fortunato dela pea susunod kahi...        3.0  \n",
       "418  president duterte reduced changed way life eve...        7.0  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1368x792 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = df.plot(\n",
    "    x='date', \n",
    "    y='date_diff', \n",
    "    rot=0,\n",
    "    figsize=(19,11)\n",
    ").set_title(\"Differences in dates of speeches\", fontsize=15)\n",
    "plt.xticks(fontsize=9)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# N-gram and TF-IDF"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def texts_to_ngram(text, n=2):\n",
    "    vectorizer = CountVectorizer(\n",
    "        ngram_range=(n, n),\n",
    "        token_pattern=r'\\b\\w+\\b',\n",
    "        min_df=1,\n",
    "        max_features=2000)\n",
    "    dtm = vectorizer.fit_transform(text)\n",
    "    #pickle.dump(vectorizer, open('out/ngram.pk', 'wb'))\n",
    "    return dtm, vectorizer\n",
    "\n",
    "def texts_to_tfidf(text):\n",
    "    vectorizer = TfidfVectorizer(max_features=2000)\n",
    "    tfidf = vectorizer.fit_transform(text)\n",
    "    #pickle.dump(vectorizer, open('out/tfidf.pk', 'wb'))\n",
    "    return tfidf, vectorizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 327,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Ngram matrix shape: (419, 2000)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[('kindly sit thank', 766),\n",
       " ('sit thank courtesy', 1637),\n",
       " ('give tikas pahinga', 387),\n",
       " ('vice president maria', 1789),\n",
       " ('president maria leonor', 1406),\n",
       " ('national defense secretary', 1156),\n",
       " ('defense secretary delfin', 216),\n",
       " ('secretary delfin lorenzana', 1567),\n",
       " ('delfin lorenzana members', 221),\n",
       " ('lorenzana members cabinet', 905)]"
      ]
     },
     "execution_count": 327,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Get ngram matrix\n",
    "ngram, ngram_v = texts_to_ngram(df['clean_text'], n=3)\n",
    "print('Ngram matrix shape:', ngram.toarray().shape)\n",
    "list(ngram_v.vocabulary_.items())[0:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 335,
   "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>title</th>\n",
       "      <th>date</th>\n",
       "      <th>text</th>\n",
       "      <th>clean_text</th>\n",
       "      <th>date_diff</th>\n",
       "      <th>swear_words_num</th>\n",
       "      <th>swear_words</th>\n",
       "      <th>modified_text</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>Speech of President Rodrigo Roa Duterte during...</td>\n",
       "      <td>2018-03-18</td>\n",
       "      <td>\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...</td>\n",
       "      <td>kindly sit thank courtesy troop commander coul...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>5</td>\n",
       "      <td>[g***, b***, b***, g***, t**** i**]</td>\n",
       "      <td>Kindly sit down. Thank you for your courtesy. ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>Speech of President Rodrigo Roa Duterte during...</td>\n",
       "      <td>2018-03-20</td>\n",
       "      <td>\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...</td>\n",
       "      <td>kindly sit salamat ho kortesiya ninyo troop co...</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0</td>\n",
       "      <td>[]</td>\n",
       "      <td>Kindly sit down and salamat ho sa kortesiya ni...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>Speech of President Rodrigo Roa Duterte during...</td>\n",
       "      <td>2018-03-20</td>\n",
       "      <td>\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...</td>\n",
       "      <td>kindly sit thank courtesy ladies gentlemen usu...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>51</td>\n",
       "      <td>[p***** i**, p****** i****, t***, l****, g***,...</td>\n",
       "      <td>Kindly sit down. Thank you for your courtesy, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>Speech of President Rodrigo Roa Duterte during...</td>\n",
       "      <td>2018-03-21</td>\n",
       "      <td>\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...</td>\n",
       "      <td>kindly sit thank courtesy troop commander coul...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>17</td>\n",
       "      <td>[g***, g***, b****, p***** i**, l****, p***** ...</td>\n",
       "      <td>Kindly sit down. Thank you for your courtesy. ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>Speech of President Rodrigo Roa Duterte during...</td>\n",
       "      <td>2018-03-21</td>\n",
       "      <td>\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...</td>\n",
       "      <td>duterte duterte hinanap kayo noong election ni...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>82</td>\n",
       "      <td>[l****, l*****, g***, p**** i***, p***** i**, ...</td>\n",
       "      <td>Duterte, Duterte, hinanap ko kayo noong electi...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                               title       date  \\\n",
       "0  Speech of President Rodrigo Roa Duterte during... 2018-03-18   \n",
       "1  Speech of President Rodrigo Roa Duterte during... 2018-03-20   \n",
       "2  Speech of President Rodrigo Roa Duterte during... 2018-03-20   \n",
       "3  Speech of President Rodrigo Roa Duterte during... 2018-03-21   \n",
       "4  Speech of President Rodrigo Roa Duterte during... 2018-03-21   \n",
       "\n",
       "                                                text  \\\n",
       "0  \\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...   \n",
       "1  \\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...   \n",
       "2  \\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...   \n",
       "3  \\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...   \n",
       "4  \\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...   \n",
       "\n",
       "                                          clean_text  date_diff  \\\n",
       "0  kindly sit thank courtesy troop commander coul...        NaN   \n",
       "1  kindly sit salamat ho kortesiya ninyo troop co...        2.0   \n",
       "2  kindly sit thank courtesy ladies gentlemen usu...        0.0   \n",
       "3  kindly sit thank courtesy troop commander coul...        1.0   \n",
       "4  duterte duterte hinanap kayo noong election ni...        0.0   \n",
       "\n",
       "   swear_words_num                                        swear_words  \\\n",
       "0                5                [g***, b***, b***, g***, t**** i**]   \n",
       "1                0                                                 []   \n",
       "2               51  [p***** i**, p****** i****, t***, l****, g***,...   \n",
       "3               17  [g***, g***, b****, p***** i**, l****, p***** ...   \n",
       "4               82  [l****, l*****, g***, p**** i***, p***** i**, ...   \n",
       "\n",
       "                                       modified_text  \n",
       "0  Kindly sit down. Thank you for your courtesy. ...  \n",
       "1  Kindly sit down and salamat ho sa kortesiya ni...  \n",
       "2  Kindly sit down. Thank you for your courtesy, ...  \n",
       "3  Kindly sit down. Thank you for your courtesy. ...  \n",
       "4  Duterte, Duterte, hinanap ko kayo noong electi...  "
      ]
     },
     "execution_count": 335,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Modify original text to cut out first part and to remove line breaks\n",
    "def modify_data(text):\n",
    "    if not isinstance(text, float):\n",
    "        index_of_date = text.find(']')\n",
    "        text = text[index_of_date+1:]\n",
    "        text = ' '.join(text.split())\n",
    "    \n",
    "    return text\n",
    "\n",
    "df['modified_text'] = df['text'].apply(modify_data)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TFIDF matrix shape: (419, 2000)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[('kindly', 899),\n",
       " ('sit', 1673),\n",
       " ('thank', 1806),\n",
       " ('courtesy', 368),\n",
       " ('commander', 329),\n",
       " ('could', 361),\n",
       " ('give', 640),\n",
       " ('order', 1315),\n",
       " ('vice', 1923),\n",
       " ('president', 1448)]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Get TFIDF matrix\n",
    "tfidf, tfidf_v = texts_to_tfidf(df['clean_text'])\n",
    "print('TFIDF matrix shape:', tfidf.toarray().shape)\n",
    "list(tfidf_v.vocabulary_.items())[0:10]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Term Frequencies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "def vector_to_frequency(vector, vectorizer):\n",
    "    \"\"\"\n",
    "    Return a list of words and their corresponding occurence in the corpus\n",
    "    \"\"\"\n",
    "    total = vector.sum(axis=0)\n",
    "    frequency = [(w, total[0, i]) for w, i in vectorizer.vocabulary_.items()]\n",
    "    frequency = pd.DataFrame(frequency, columns=['term', 'frequency'])\n",
    "    frequency = frequency.sort_values(by='frequency', ascending=False).reset_index(drop=True)\n",
    "    return frequency"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 332,
   "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>term</th>\n",
       "      <th>frequency</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>mga p i</td>\n",
       "      <td>270</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>p i mo</td>\n",
       "      <td>219</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>sir president duterte</td>\n",
       "      <td>204</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>p i ka</td>\n",
       "      <td>175</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>p i yan</td>\n",
       "      <td>136</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    term  frequency\n",
       "0                mga p i        270\n",
       "1                 p i mo        219\n",
       "2  sir president duterte        204\n",
       "3                 p i ka        175\n",
       "4                p i yan        136"
      ]
     },
     "execution_count": 332,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "freq_ngram = vector_to_frequency(ngram, ngram_v)\n",
    "freq_ngram.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 334,
   "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>term</th>\n",
       "      <th>frequency</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>yan</td>\n",
       "      <td>61.588303</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>mga</td>\n",
       "      <td>60.735861</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>mo</td>\n",
       "      <td>57.413046</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>hindi</td>\n",
       "      <td>47.479250</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>eh</td>\n",
       "      <td>35.609758</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    term  frequency\n",
       "0    yan  61.588303\n",
       "1    mga  60.735861\n",
       "2     mo  57.413046\n",
       "3  hindi  47.479250\n",
       "4     eh  35.609758"
      ]
     },
     "execution_count": 334,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "freq_tfidf = vector_to_frequency(tfidf, tfidf_v)\n",
    "freq_tfidf.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Plotting Frequencies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1800x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = freq_ngram.head(20).plot.bar(\n",
    "    x='term', \n",
    "    y='frequency', \n",
    "    rot=0,\n",
    "    figsize=(25,8)\n",
    ").set_title(\"Frequent terms in N-gram\", fontsize=15)\n",
    "plt.xticks(fontsize=9)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1800x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax2 = freq_tfidf.head(20).plot.bar(\n",
    "    x='term', \n",
    "    y='frequency', \n",
    "    rot=0,\n",
    "    figsize=(25,8)\n",
    ").set_title(\"Frequent terms in N-gram\", fontsize=15)\n",
    "plt.xticks(fontsize=12)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Link: https://github.com/daniddelrio/Philippine-Contraception-Discourse/blob/master/05_text_analysis_comments.ipynb"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Swear Words per Speech"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "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>title</th>\n",
       "      <th>date</th>\n",
       "      <th>text</th>\n",
       "      <th>clean_text</th>\n",
       "      <th>date_diff</th>\n",
       "      <th>swear_words_num</th>\n",
       "      <th>swear_words</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>Speech of President Rodrigo Roa Duterte during...</td>\n",
       "      <td>2018-03-18</td>\n",
       "      <td>\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...</td>\n",
       "      <td>kindly sit thank courtesy troop commander coul...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>5</td>\n",
       "      <td>[g***, b***, b***, g***, t**** i**]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>Speech of President Rodrigo Roa Duterte during...</td>\n",
       "      <td>2018-03-20</td>\n",
       "      <td>\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...</td>\n",
       "      <td>kindly sit salamat ho kortesiya ninyo troop co...</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0</td>\n",
       "      <td>[]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>Speech of President Rodrigo Roa Duterte during...</td>\n",
       "      <td>2018-03-20</td>\n",
       "      <td>\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...</td>\n",
       "      <td>kindly sit thank courtesy ladies gentlemen usu...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>51</td>\n",
       "      <td>[p***** i**, p****** i****, t***, l****, g***,...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>Speech of President Rodrigo Roa Duterte during...</td>\n",
       "      <td>2018-03-21</td>\n",
       "      <td>\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...</td>\n",
       "      <td>kindly sit thank courtesy troop commander coul...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>17</td>\n",
       "      <td>[g***, g***, b****, p***** i**, l****, p***** ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>Speech of President Rodrigo Roa Duterte during...</td>\n",
       "      <td>2018-03-21</td>\n",
       "      <td>\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...</td>\n",
       "      <td>duterte duterte hinanap kayo noong election ni...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>82</td>\n",
       "      <td>[l****, l*****, g***, p**** i***, p***** i**, ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                               title       date  \\\n",
       "0  Speech of President Rodrigo Roa Duterte during... 2018-03-18   \n",
       "1  Speech of President Rodrigo Roa Duterte during... 2018-03-20   \n",
       "2  Speech of President Rodrigo Roa Duterte during... 2018-03-20   \n",
       "3  Speech of President Rodrigo Roa Duterte during... 2018-03-21   \n",
       "4  Speech of President Rodrigo Roa Duterte during... 2018-03-21   \n",
       "\n",
       "                                                text  \\\n",
       "0  \\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...   \n",
       "1  \\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...   \n",
       "2  \\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...   \n",
       "3  \\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...   \n",
       "4  \\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n...   \n",
       "\n",
       "                                          clean_text  date_diff  \\\n",
       "0  kindly sit thank courtesy troop commander coul...        NaN   \n",
       "1  kindly sit salamat ho kortesiya ninyo troop co...        2.0   \n",
       "2  kindly sit thank courtesy ladies gentlemen usu...        0.0   \n",
       "3  kindly sit thank courtesy troop commander coul...        1.0   \n",
       "4  duterte duterte hinanap kayo noong election ni...        0.0   \n",
       "\n",
       "   swear_words_num                                        swear_words  \n",
       "0                5                [g***, b***, b***, g***, t**** i**]  \n",
       "1                0                                                 []  \n",
       "2               51  [p***** i**, p****** i****, t***, l****, g***,...  \n",
       "3               17  [g***, g***, b****, p***** i**, l****, p***** ...  \n",
       "4               82  [l****, l*****, g***, p**** i***, p***** i**, ...  "
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "swear_words_pattern = r'([a-z]\\*{2,} [a-z]\\*{2,}|[a-z]\\*{2,})'\n",
    "df['swear_words_num'] = df['clean_text'].apply(lambda x : len(re.findall(swear_words_pattern, x)))\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x12342c910>"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.violinplot(data=df, x='swear_words_num')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[<matplotlib.axes._subplots.AxesSubplot object at 0x12b9a2510>]],\n",
       "      dtype=object)"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.hist(column='swear_words_num', bins=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'p***** i**': 1710,\n",
       " 'y***': 406,\n",
       " 'g***': 388,\n",
       " 't*** i**': 386,\n",
       " 'p***': 339,\n",
       " 'p***** i****': 334,\n",
       " 'l****': 212,\n",
       " 'p****** i**': 186,\n",
       " 'b****': 182,\n",
       " 's***': 170}"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['swear_words'] = df['clean_text'].apply(lambda x : re.findall(swear_words_pattern, x))\n",
    "swear_words_freq = {}\n",
    "for i in range(df.shape[0]):\n",
    "    for word in df.loc[i, 'swear_words']:\n",
    "        if word not in swear_words_freq:\n",
    "            swear_words_freq[word] = 1\n",
    "        else:\n",
    "            swear_words_freq[word] += 1\n",
    "\n",
    "swear_words_freq = {k: v for k, v in sorted(swear_words_freq.items(), \n",
    "                                            key=lambda item: item[1],\n",
    "                                            reverse=True)[:10]}\n",
    "swear_words_freq"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Proper Nouns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 316,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('Vice President Maria Leonor Robredo',\n",
       "  'Vice President Maria Leonor Robredo',\n",
       "  ' Robredo'),\n",
       " ('e National', '', ''),\n",
       " ('Defense Secretary Delfin Lorenzana',\n",
       "  'Defense Secretary Delfin Lorenzana',\n",
       "  ' Lorenzana'),\n",
       " ('e Cabinet', '', ''),\n",
       " ('AFP Chief', 'AFP Chief', ' Chief'),\n",
       " ('f Staff', '', ''),\n",
       " ('General Rey Leonardo Guerrero',\n",
       "  'General Rey Leonardo Guerrero',\n",
       "  ' Guerrero'),\n",
       " ('e Philippine', '', ''),\n",
       " ('Military Academy', 'Military Academy', ' Academy'),\n",
       " ('PMA Superintendent', 'PMA Superintendent', ' Superintendent')]"
      ]
     },
     "execution_count": 316,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# proper_nouns_pattern = r'[A-za-z] [A-Z]\\w+'\n",
    "proper_nouns_pattern = r'(([A-Z]\\w+( [A-Z]\\w+)+)|[A-Za-z] [A-Z]\\w+(?:-\\w+)*)'\n",
    "# proper_nouns_pattern_2 = r'[A-Za-z]+-[A-Z][A-Za-z]+'\n",
    "pattern = r'[A-Za-z]+-*[A-za-z]+'\n",
    "\n",
    "sample = df.loc[0, 'text']\n",
    "index_of_date = sample.find(']')\n",
    "sample = sample[index_of_date+1:]\n",
    "re.findall(proper_nouns_pattern, sample)[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 323,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Davao': 1724,\n",
       " 'PRESIDENT DUTERTE': 1320,\n",
       " 'Philippines': 1224,\n",
       " 'Mindanao': 814,\n",
       " 'President': 756,\n",
       " 'God': 736,\n",
       " 'Cabinet': 675,\n",
       " 'Pilipino': 674,\n",
       " 'Filipino': 662,\n",
       " 'China': 654,\n",
       " 'NPA': 493,\n",
       " 'Bong': 468,\n",
       " 'Pilipinas': 446,\n",
       " 'Secretary': 439,\n",
       " 'Filipinos': 411,\n",
       " 'Duterte': 381,\n",
       " 'Congress': 368,\n",
       " 'Moro': 356,\n",
       " 'Inday': 322,\n",
       " 'Manila': 318,\n",
       " 'Bisaya': 317,\n",
       " 'America': 310,\n",
       " 'Philippine': 304,\n",
       " 'Bato': 301,\n",
       " 'Presidente': 299,\n",
       " 'Diyos': 294,\n",
       " 'General': 290,\n",
       " 'Boracay': 288,\n",
       " 'Cebu': 279,\n",
       " 'Chinese': 276}"
      ]
     },
     "execution_count": 323,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "proper_nouns_freq = {}\n",
    "for i in range(df.shape[0]):\n",
    "    text = df.loc[i, 'text']\n",
    "    index_of_date = text.find(']')\n",
    "    text = text[index_of_date+1:]\n",
    "    proper_nouns = re.findall(proper_nouns_pattern, text)\n",
    "    proper_nouns = [x[0][2:] if x[0][1] == ' ' else x[0] for x in proper_nouns]\n",
    "    for word in proper_nouns:\n",
    "        if word not in proper_nouns_freq:\n",
    "            proper_nouns_freq[word] = 1\n",
    "        else:\n",
    "            proper_nouns_freq[word] += 1\n",
    "\n",
    "proper_nouns_freq = {k: v for k, v in sorted(proper_nouns_freq.items(), \n",
    "                                            key=lambda item: item[1],\n",
    "                                            reverse=True)[:30]}\n",
    "proper_nouns_freq"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# PyLDAVis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 339,
   "metadata": {},
   "outputs": [],
   "source": [
    "pyLDAvis.save_html(data=df['clean_text'],fileobj='pyldavis.html')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# WordCloud"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 343,
   "metadata": {},
   "outputs": [
    {
     "ename": "TypeError",
     "evalue": "__init__() got an unexpected keyword argument 'words'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-343-2ea43408fe46>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mWordCloud\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mwords\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdf\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'clean_text'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;31mTypeError\u001b[0m: __init__() got an unexpected keyword argument 'words'"
     ]
    }
   ],
   "source": [
    "WordCloud()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "dutertespeech",
   "language": "python",
   "name": "dutertespeech"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}