Python/state_of_union_v2.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Analyze state of the union addresses. \n",
    "Data source: https://en.wikisource.org/wiki/Portal:State_of_the_Union_Speeches_by_United_States_Presidents\n",
    "scrape the text of all speeches and then maybe try to find patterns of speech of each president?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "https://nlp.stanford.edu/IR-book/html/htmledition/sublinear-tf-scaling-1.html \n",
    "https://www.oreilly.com/library/view/applied-text-analysis/9781491963036/ch04.html "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Setup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "from bs4 import BeautifulSoup\n",
    "import requests\n",
    "import pandas as pd\n",
    "import urllib.request\n",
    "import re\n",
    "\n",
    "from textblob import TextBlob\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "from gensim.corpora.dictionary import Dictionary\n",
    "import gensim\n",
    "from gensim import corpora, models\n",
    "import spacy\n",
    "import time\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer, CountVectorizer\n",
    "from sklearn.decomposition import LatentDirichletAllocation\n",
    "\n",
    "# Data source we are going to scrape for results\n",
    "data_url = 'https://en.wikisource.org/wiki/Portal:State_of_the_Union_Speeches_by_United_States_Presidents'\n",
    "\n",
    "link_list = []\n",
    "\n",
    "# extract the text of a speech from a URL\n",
    "# text is extracted in a list of paragraphs (strings) for each speech\n",
    "def get_speech(url):\n",
    "    return([ p.text.strip() for p in BeautifulSoup(urllib.request.urlopen(url)).find_all(\"p\") if \\\n",
    "             'This work is in the public domain in the United States because it is a work of the United States federal government' \\\n",
    "             not in p.text.strip()])\n",
    "\n",
    "# Make a frequency count by distinct values of \n",
    "# column(s) listed in 'groupbyvars'\n",
    "# Returns pandas dataframe\n",
    "def tidy_count(df,groupbyvars):\n",
    "    return(df.groupby(groupbyvars).size().reset_index().\\\n",
    "        rename(columns={0: \"n\"}).sort_values('n',ascending=False).reset_index(drop=True))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Web Scraping"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "resp = urllib.request.urlopen(data_url)\n",
    "soup = BeautifulSoup(resp, from_encoding=resp.info().get_param('charset'))\n",
    "\n",
    "# Get all links to state of the union addresses from \n",
    "for link in soup.find_all('a', href=True):\n",
    "    if \"union_address\" in link['href'].lower() or \"union_speech\" in link['href'].lower() \\\n",
    "        and \"portal\" not in link['href'].lower() and \"#\" not in link['href'].lower():\n",
    "        link_list.append(link['href'])\n",
    "\n",
    "# Note that I am storing these speeches as lists of paragraphs (strings) for readability\n",
    "speeches = [get_speech('https://en.wikisource.org' + link) for link in link_list]\n",
    "# Extract presidents names from link text\n",
    "presidents = [ link.replace('%','/').split('/')[2].replace('_',' ') for link in link_list ]\n",
    "\n",
    "# Extract state of the union text entries so we can extract the date\n",
    "sou_entries = []\n",
    "for item in soup.find_all('li'):\n",
    "    if 'union' in item.text.strip().lower() and '(' in  item.text.strip().lower():\n",
    "        sou_entries.append(item.text.strip())\n",
    "\n",
    "speeches_pd = pd.DataFrame({\n",
    "                'president' : presidents,\n",
    "                'speech' : speeches,\n",
    "                'year' : [int(re.findall('\\d+',item)[1]) for item in sou_entries ]} )\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>president</th>\n",
       "      <th>speech</th>\n",
       "      <th>year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>134</th>\n",
       "      <td>Calvin Coolidge</td>\n",
       "      <td>[Since the close of the last Congress the Nati...</td>\n",
       "      <td>1923</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>169</th>\n",
       "      <td>Dwight D. Eisenhower</td>\n",
       "      <td>[Mr. President, Mr. Speaker, Members of the 85...</td>\n",
       "      <td>1958</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>81</th>\n",
       "      <td>Ulysses S. Grant</td>\n",
       "      <td>[To the Senate and House of Representatives:, ...</td>\n",
       "      <td>1870</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>George Washington</td>\n",
       "      <td>[I embrace with great satisfaction the opportu...</td>\n",
       "      <td>1790</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>198</th>\n",
       "      <td>Ronald Reagan</td>\n",
       "      <td>[Mr. Speaker, Mr. President, distinguished mem...</td>\n",
       "      <td>1986</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                president                                             speech  \\\n",
       "134       Calvin Coolidge  [Since the close of the last Congress the Nati...   \n",
       "169  Dwight D. Eisenhower  [Mr. President, Mr. Speaker, Members of the 85...   \n",
       "81       Ulysses S. Grant  [To the Senate and House of Representatives:, ...   \n",
       "0       George Washington  [I embrace with great satisfaction the opportu...   \n",
       "198         Ronald Reagan  [Mr. Speaker, Mr. President, distinguished mem...   \n",
       "\n",
       "     year  \n",
       "134  1923  \n",
       "169  1958  \n",
       "81   1870  \n",
       "0    1790  \n",
       "198  1986  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "speeches_pd.sample(n=5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preprocessing"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Clean text (remove stop words, convert to lower case, remove non-alphabetic text, and lemmatize"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "nlp = spacy.load('en', disable=['parser', 'ner']) # disabling the parser makes it run faster\n",
    "#nlp = spacy.load('en_core_web_lg',disable=['parser', 'ner'])\n",
    "\n",
    "# Preprocess pipeline with spacy. \n",
    "def spacy_preprocess(text): \n",
    "    text_out = []\n",
    "    for token in nlp(text.lower()):\n",
    "        # Get rid of stop words and non-alphanumeric\n",
    "        if not token.is_stop and token.is_alpha and len(token)>1:\n",
    "            lemma = token.lemma_\n",
    "            if lemma != '-PRON-':\n",
    "                text_out.append(lemma)\n",
    "            else:\n",
    "                text_out.append(token.text)            \n",
    "    return(text_out)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "True\n",
      "False\n"
     ]
    }
   ],
   "source": [
    "# Case sensitivity matters for stop-words for some reason\n",
    "# So we need to convert all text to lowercase before\n",
    "# letting spacy see it (as done in above function)\n",
    "for token in nlp('The the'):\n",
    "    print(not token.is_stop)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['dog', 'run', 'bob', 'see', 'squirrel']"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# test spacy preprocessing\n",
    "spacy_preprocess('The dog ran into Bob beCause he saw 234 squirrels under VAU15')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Preprocessing time elapsed: 100.60255813598633\n"
     ]
    }
   ],
   "source": [
    "# Each speech is stored as a list of paragraph strings. \n",
    "# Here we join the paragraphs into a single speech string\n",
    "speech_list = [\" \".join(speech) for speech in speeches_pd['speech'].tolist() ]\n",
    "\n",
    "## Pre-process and tokenize our speeches\n",
    "start = time.time()\n",
    "tokens = [spacy_preprocess(speech) for \\\n",
    "          speech in speech_list ]\n",
    "print('Preprocessing time elapsed: ' + str(time.time()-start))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Gensim Topic Modeling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# create dictionary and corpus objects for gensim topic modeling\n",
    "common_dictionary = Dictionary(tokens)\n",
    "gensim_corpus = [common_dictionary.doc2bow(doc) for doc in tokens]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Gensim LDA training time: 67.68750023841858\n"
     ]
    }
   ],
   "source": [
    "start = time.time()\n",
    "gensimLDA = gensim.models.ldamodel.LdaModel(gensim_corpus,num_topics=12,\n",
    "        random_state=42, id2word=common_dictionary,passes=10)\n",
    "print('Gensim LDA training time: ' + str(time.time()-start))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "topic 0: government man state law country nation american people great good united power year right public case work matter\n",
      "topic 1: government state united year congress nation country time great people shall law american good act war program interest\n",
      "topic 2: american year america child work people new tonight school nation family good congress world help country ask know\n",
      "topic 3: year american people new nation world america government work congress program state need time good help war country\n",
      "topic 4: government year congress federal country state public increase department national service law expenditure work system large need present\n",
      "topic 5: state government united congress great country public power war people constitution year law time present act duty right\n",
      "topic 6: year state government congress people new work nation american country time great child united law america act need\n",
      "topic 7: state year united government american nation law country congress people new work great public time service need power\n",
      "topic 8: state government united year congress country law great public present interest subject time duty general power act department\n",
      "topic 9: nation great government people man world good need law work congress national country state power year time war\n",
      "topic 10: war enemy british state government united american country prisoner savage year nation great good force arm time power\n",
      "topic 11: great enemy american state war government country force public time congress united people year lake command service peace\n"
     ]
    }
   ],
   "source": [
    "# Print the words in each gensim topic without probabilties\n",
    "# to show probabilites, just use the gensimModel.show_topics() function\n",
    "def print_gensim_topics(gensimModel,n_words):\n",
    "    gensim_topic_list = gensimModel.print_topics(num_words=n_words)\n",
    "    for n, topic in enumerate(gensim_topic_list):\n",
    "        # Grab words that are in double quotes (ie. strip out probabilities)\n",
    "        print('topic ' + str(n) + \": \" + \" \".join(re.findall(r'\"(.*?)\"', topic[1])))\n",
    "\n",
    "print_gensim_topics(gensimLDA,18)\n",
    "#gensimLDA.show_topics()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "doc_top_proportions = [ gensimLDA.get_document_topics(doc,minimum_probability=0) for doc in gensim_corpus ]\n",
    "doc_top_probs = pd.DataFrame([[prob for (topic, prob) in prob_tuple] for prob_tuple in doc_top_proportions])\n",
    "doc_top_probs['MostCommonTopic'] = doc_top_probs.idxmax(axis=1)\n",
    "doc_top_probs[['year','president']] = speeches_pd[['year','president']]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Topic 10 seems to be related to war and peace, plot the prevalence of the topic"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x7f6bf1125510>"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 648x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.relplot(x=\"year\",y=10, data=doc_top_probs,kind=\"line\",height=6,aspect=1.5)\n",
    "#plt.xlim(1995,2018)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "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>MostCommonTopic</th>\n",
       "      <th>n</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>8</td>\n",
       "      <td>95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3</td>\n",
       "      <td>63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>9</td>\n",
       "      <td>31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>4</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   MostCommonTopic   n\n",
       "0                8  95\n",
       "1                3  63\n",
       "2                9  31\n",
       "3                5  20\n",
       "4                2  12\n",
       "5                4   9\n",
       "6                0   1\n",
       "7               10   1\n",
       "8               11   1"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tidy_count(doc_top_probs,'MostCommonTopic')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "      <th>MostCommonTopic</th>\n",
       "      <th>year</th>\n",
       "      <th>president</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>50</th>\n",
       "      <td>0.000015</td>\n",
       "      <td>0.000015</td>\n",
       "      <td>0.000015</td>\n",
       "      <td>0.000015</td>\n",
       "      <td>0.000015</td>\n",
       "      <td>0.818892</td>\n",
       "      <td>0.000015</td>\n",
       "      <td>0.000015</td>\n",
       "      <td>0.180957</td>\n",
       "      <td>0.000015</td>\n",
       "      <td>0.000015</td>\n",
       "      <td>0.000015</td>\n",
       "      <td>5</td>\n",
       "      <td>1839</td>\n",
       "      <td>Martin Van Buren</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>154</th>\n",
       "      <td>0.000056</td>\n",
       "      <td>0.000056</td>\n",
       "      <td>0.000056</td>\n",
       "      <td>0.685221</td>\n",
       "      <td>0.000056</td>\n",
       "      <td>0.059241</td>\n",
       "      <td>0.000056</td>\n",
       "      <td>0.000056</td>\n",
       "      <td>0.002226</td>\n",
       "      <td>0.252861</td>\n",
       "      <td>0.000056</td>\n",
       "      <td>0.000056</td>\n",
       "      <td>3</td>\n",
       "      <td>1944</td>\n",
       "      <td>Franklin Delano Roosevelt</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>67</th>\n",
       "      <td>0.000019</td>\n",
       "      <td>0.000019</td>\n",
       "      <td>0.000019</td>\n",
       "      <td>0.000019</td>\n",
       "      <td>0.000019</td>\n",
       "      <td>0.990903</td>\n",
       "      <td>0.000019</td>\n",
       "      <td>0.000019</td>\n",
       "      <td>0.008912</td>\n",
       "      <td>0.000019</td>\n",
       "      <td>0.000019</td>\n",
       "      <td>0.000019</td>\n",
       "      <td>5</td>\n",
       "      <td>1856</td>\n",
       "      <td>Franklin Pierce</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>161</th>\n",
       "      <td>0.000048</td>\n",
       "      <td>0.000048</td>\n",
       "      <td>0.000048</td>\n",
       "      <td>0.443746</td>\n",
       "      <td>0.000048</td>\n",
       "      <td>0.000048</td>\n",
       "      <td>0.000048</td>\n",
       "      <td>0.000048</td>\n",
       "      <td>0.000048</td>\n",
       "      <td>0.555775</td>\n",
       "      <td>0.000048</td>\n",
       "      <td>0.000048</td>\n",
       "      <td>9</td>\n",
       "      <td>1951</td>\n",
       "      <td>Harry S. Truman</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>0.000062</td>\n",
       "      <td>0.000062</td>\n",
       "      <td>0.000062</td>\n",
       "      <td>0.000062</td>\n",
       "      <td>0.000062</td>\n",
       "      <td>0.271553</td>\n",
       "      <td>0.000062</td>\n",
       "      <td>0.000062</td>\n",
       "      <td>0.727831</td>\n",
       "      <td>0.000062</td>\n",
       "      <td>0.000062</td>\n",
       "      <td>0.000062</td>\n",
       "      <td>8</td>\n",
       "      <td>1815</td>\n",
       "      <td>James Madison</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            0         1         2         3         4         5         6  \\\n",
       "50   0.000015  0.000015  0.000015  0.000015  0.000015  0.818892  0.000015   \n",
       "154  0.000056  0.000056  0.000056  0.685221  0.000056  0.059241  0.000056   \n",
       "67   0.000019  0.000019  0.000019  0.000019  0.000019  0.990903  0.000019   \n",
       "161  0.000048  0.000048  0.000048  0.443746  0.000048  0.000048  0.000048   \n",
       "26   0.000062  0.000062  0.000062  0.000062  0.000062  0.271553  0.000062   \n",
       "\n",
       "            7         8         9        10        11  MostCommonTopic  year  \\\n",
       "50   0.000015  0.180957  0.000015  0.000015  0.000015                5  1839   \n",
       "154  0.000056  0.002226  0.252861  0.000056  0.000056                3  1944   \n",
       "67   0.000019  0.008912  0.000019  0.000019  0.000019                5  1856   \n",
       "161  0.000048  0.000048  0.555775  0.000048  0.000048                9  1951   \n",
       "26   0.000062  0.727831  0.000062  0.000062  0.000062                8  1815   \n",
       "\n",
       "                     president  \n",
       "50            Martin Van Buren  \n",
       "154  Franklin Delano Roosevelt  \n",
       "67             Franklin Pierce  \n",
       "161            Harry S. Truman  \n",
       "26               James Madison  "
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "doc_top_probs.sample(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "#doc_top_probs.drop('MostCommonTopic', axis=1).idxmax(axis=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## SkLearn Topic Modeling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Vectorize\n",
    "#tvec = TfidfVectorizer(tokenizer=None,analyzer='word',\n",
    "#                        max_df=0.95, min_df=2,\n",
    "#                        ngram_range=(1,1),max_features=10000)\n",
    "\n",
    "tf_vectorizer = CountVectorizer(max_df=0.95, min_df=2)\n",
    "                        #        max_features=10000)\n",
    "tf = pd.DataFrame(tf_vectorizer.fit_transform([\" \".join(doc) for doc in tokens]).toarray(),\n",
    "                     columns  = tf_vectorizer.get_feature_names())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(233, 11440)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sklearn LDA training time: 49.22464942932129\n"
     ]
    }
   ],
   "source": [
    "## LDA Topic Modeling\n",
    "speech_lda = LatentDirichletAllocation(n_components=8,\n",
    "                                       random_state=42,\n",
    "                                    #   evaluate_every=0.5, # eval perplexity\n",
    "                                    #   perp_tol=0.05,\n",
    "                                       max_iter=30,\n",
    "                                learning_method='online'\n",
    "                                      )\n",
    "start = time.time()\n",
    "speech_lda.fit(tf) # fit model\n",
    "print('Sklearn LDA training time: ' + str(time.time()-start))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Topic 0: present increase department service general report subject american large duty secretary result citizen condition foreign land shall treaty legislation recommend\n",
      "Topic 1: american work increase america need system world present business citizen department service result report subject program job tax large foreign\n",
      "Topic 2: present citizen duty general secretary service report necessary treaty subject increase number condition department large shall american work purpose foreign\n",
      "Topic 3: american world america work program need help federal tax economic security job increase million know child economy free let life\n",
      "Topic 4: present work increase man duty treaty american department general shall subject service result legislation question policy office attention land commerce\n",
      "Topic 5: man work need business condition american labor present service necessary matter far possible way increase corporation action shall large come\n",
      "Topic 6: present citizen treaty duty subject general service department report treasury foreign system revenue increase question mexico work land man necessary\n",
      "Topic 7: duty present citizen subject treaty shall constitution object mexico union general effect necessary commerce territory treasury measure receive place require\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Print the top words for topics in the topic model\n",
    "def print_top_words(model, feature_names, n_top_words):\n",
    "    for topic_idx, topic in enumerate(model.components_):\n",
    "        message = \"Topic %d: \" % topic_idx\n",
    "        message += \" \".join([feature_names[i]\n",
    "                             for i in topic.argsort()[:-n_top_words - 1:-1]])\n",
    "        print(message)\n",
    "    print()    \n",
    "\n",
    "print_top_words(speech_lda, tf.columns.values, 20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "# % makeup of each document by topic\n",
    "doc_topic = pd.DataFrame(speech_lda.transform(tf),\n",
    "                        columns=list(range(0,(speech_lda.n_components))))\n",
    "\n",
    "# What was the most common topic in each document?\n",
    "doc_topic['MostCommonTopic'] = doc_topic.idxmax(axis=1)\n",
    "doc_topic[['year','president']] = speeches_pd[['year','president']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "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>MostCommonTopic</th>\n",
       "      <th>n</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>90</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>7</td>\n",
       "      <td>70</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>58</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   MostCommonTopic   n\n",
       "0                3  90\n",
       "1                7  70\n",
       "2                0  58\n",
       "3                5  15"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tidy_count(doc_topic,'MostCommonTopic')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0                  105\n",
       "1                    0\n",
       "2                    0\n",
       "3                  212\n",
       "4                    0\n",
       "5                  119\n",
       "6                    0\n",
       "7                   35\n",
       "MostCommonTopic      0\n",
       "year               232\n",
       "dtype: int64"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# In which document did each topic most commonly occur?\n",
    "doc_topic.drop('president',axis=1).idxmax(axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "import seaborn as sns\n",
    "cm = sns.light_palette(\"green\", as_cmap=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "president_topic = doc_topic.drop(['MostCommonTopic','year'],axis=1).groupby('president').agg('mean')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "    #T_e2291480_6319_11eb_97c0_39a9936bdd38row0_col0 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row0_col1 {\n",
       "            background-color:  #cbe4cb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row0_col2 {\n",
       "            background-color:  #cbe4cb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row0_col3 {\n",
       "            background-color:  #008000;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row0_col4 {\n",
       "            background-color:  #cbe4cb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row0_col5 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row0_col6 {\n",
       "            background-color:  #cbe4cb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row0_col7 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row1_col0 {\n",
       "            background-color:  #eaf2ea;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row1_col1 {\n",
       "            background-color:  #128912;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row1_col2 {\n",
       "            background-color:  #128912;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row1_col3 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row1_col4 {\n",
       "            background-color:  #128912;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row1_col5 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row1_col6 {\n",
       "            background-color:  #128912;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row1_col7 {\n",
       "            background-color:  #018001;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row2_col0 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row2_col1 {\n",
       "            background-color:  #b7dab7;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row2_col2 {\n",
       "            background-color:  #b7dab7;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row2_col3 {\n",
       "            background-color:  #008000;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row2_col4 {\n",
       "            background-color:  #b7dab7;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row2_col5 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row2_col6 {\n",
       "            background-color:  #b7dab7;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row2_col7 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row3_col0 {\n",
       "            background-color:  #d6e9d6;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row3_col1 {\n",
       "            background-color:  #bfdebf;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row3_col2 {\n",
       "            background-color:  #bfdebf;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row3_col3 {\n",
       "            background-color:  #0f870f;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row3_col4 {\n",
       "            background-color:  #bfdebf;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row3_col5 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row3_col6 {\n",
       "            background-color:  #bfdebf;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row3_col7 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row4_col0 {\n",
       "            background-color:  #deedde;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row4_col1 {\n",
       "            background-color:  #a5d1a5;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row4_col2 {\n",
       "            background-color:  #a5d1a5;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row4_col3 {\n",
       "            background-color:  #0c860c;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row4_col4 {\n",
       "            background-color:  #a5d1a5;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row4_col5 {\n",
       "            background-color:  #e9f2e9;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row4_col6 {\n",
       "            background-color:  #a5d1a5;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row4_col7 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row5_col0 {\n",
       "            background-color:  #40a040;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row5_col1 {\n",
       "            background-color:  #d1e6d1;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row5_col2 {\n",
       "            background-color:  #d1e6d1;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row5_col3 {\n",
       "            background-color:  #bfdebf;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row5_col4 {\n",
       "            background-color:  #d1e6d1;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row5_col5 {\n",
       "            background-color:  #a5d1a5;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row5_col6 {\n",
       "            background-color:  #d1e6d1;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row5_col7 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row6_col0 {\n",
       "            background-color:  #008000;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row6_col1 {\n",
       "            background-color:  #bfdebf;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row6_col2 {\n",
       "            background-color:  #bfdebf;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row6_col3 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row6_col4 {\n",
       "            background-color:  #bfdebf;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row6_col5 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row6_col6 {\n",
       "            background-color:  #bfdebf;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row6_col7 {\n",
       "            background-color:  #b0d6b0;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row7_col0 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row7_col1 {\n",
       "            background-color:  #b9dbb9;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row7_col2 {\n",
       "            background-color:  #b9dbb9;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row7_col3 {\n",
       "            background-color:  #008000;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row7_col4 {\n",
       "            background-color:  #b9dbb9;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row7_col5 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row7_col6 {\n",
       "            background-color:  #b9dbb9;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row7_col7 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row8_col0 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row8_col1 {\n",
       "            background-color:  #008000;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row8_col2 {\n",
       "            background-color:  #008000;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row8_col3 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row8_col4 {\n",
       "            background-color:  #008000;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row8_col5 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row8_col6 {\n",
       "            background-color:  #008000;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row8_col7 {\n",
       "            background-color:  #008000;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row9_col0 {\n",
       "            background-color:  #86c286;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row9_col1 {\n",
       "            background-color:  #eaf2ea;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row9_col2 {\n",
       "            background-color:  #eaf2ea;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row9_col3 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row9_col4 {\n",
       "            background-color:  #eaf2ea;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row9_col5 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row9_col6 {\n",
       "            background-color:  #eaf2ea;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row9_col7 {\n",
       "            background-color:  #4ba54a;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row10_col0 {\n",
       "            background-color:  #e9f2e9;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row10_col1 {\n",
       "            background-color:  #a2cfa2;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row10_col2 {\n",
       "            background-color:  #a2cfa2;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row10_col3 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row10_col4 {\n",
       "            background-color:  #a2cfa2;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row10_col5 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row10_col6 {\n",
       "            background-color:  #a2cfa2;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row10_col7 {\n",
       "            background-color:  #018001;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row11_col0 {\n",
       "            background-color:  #349a34;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row11_col1 {\n",
       "            background-color:  #bbdcbb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row11_col2 {\n",
       "            background-color:  #bbdcbb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row11_col3 {\n",
       "            background-color:  #9acb9a;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row11_col4 {\n",
       "            background-color:  #bbdcbb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row11_col5 {\n",
       "            background-color:  #d6e9d6;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row11_col6 {\n",
       "            background-color:  #bbdcbb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row11_col7 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row12_col0 {\n",
       "            background-color:  #cfe5cf;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row12_col1 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row12_col2 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row12_col3 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row12_col4 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row12_col5 {\n",
       "            background-color:  #008000;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row12_col6 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row12_col7 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row13_col0 {\n",
       "            background-color:  #c5e0c5;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row13_col1 {\n",
       "            background-color:  #c2dfc2;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row13_col2 {\n",
       "            background-color:  #c2dfc2;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row13_col3 {\n",
       "            background-color:  #2b952b;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row13_col4 {\n",
       "            background-color:  #c2dfc2;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row13_col5 {\n",
       "            background-color:  #ddecdd;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row13_col6 {\n",
       "            background-color:  #c2dfc2;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row13_col7 {\n",
       "            background-color:  #e9f2e9;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row14_col0 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row14_col1 {\n",
       "            background-color:  #a3d0a3;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row14_col2 {\n",
       "            background-color:  #a3d0a3;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row14_col3 {\n",
       "            background-color:  #008000;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row14_col4 {\n",
       "            background-color:  #a3d0a3;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row14_col5 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row14_col6 {\n",
       "            background-color:  #a3d0a3;\n",
       "            color:  #000000;\n",
       "        }    #T_e2291480_6319_11eb_97c0_39a9936bdd38row14_col7 {\n",
       "            background-color:  #ebf3eb;\n",
       "            color:  #000000;\n",
       "        }</style><table id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38\" ><thead>    <tr>        <th class=\"blank level0\" ></th>        <th class=\"col_heading level0 col0\" >0</th>        <th class=\"col_heading level0 col1\" >1</th>        <th class=\"col_heading level0 col2\" >2</th>        <th class=\"col_heading level0 col3\" >3</th>        <th class=\"col_heading level0 col4\" >4</th>        <th class=\"col_heading level0 col5\" >5</th>        <th class=\"col_heading level0 col6\" >6</th>        <th class=\"col_heading level0 col7\" >7</th>    </tr>    <tr>        <th class=\"index_name level0\" >president</th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>    </tr></thead><tbody>\n",
       "                <tr>\n",
       "                        <th id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38level0_row0\" class=\"row_heading level0 row0\" >Bill Clinton</th>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row0_col0\" class=\"data row0 col0\" >0.000043</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row0_col1\" class=\"data row0 col1\" >0.000043</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row0_col2\" class=\"data row0 col2\" >0.000043</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row0_col3\" class=\"data row0 col3\" >0.999701</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row0_col4\" class=\"data row0 col4\" >0.000043</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row0_col5\" class=\"data row0 col5\" >0.000043</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row0_col6\" class=\"data row0 col6\" >0.000043</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row0_col7\" class=\"data row0 col7\" >0.000043</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38level0_row1\" class=\"row_heading level0 row1\" >George Washington</th>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row1_col0\" class=\"data row1 col0\" >0.005752</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row1_col1\" class=\"data row1 col1\" >0.000183</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row1_col2\" class=\"data row1 col2\" >0.000183</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row1_col3\" class=\"data row1 col3\" >0.000184</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row1_col4\" class=\"data row1 col4\" >0.000183</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row1_col5\" class=\"data row1 col5\" >0.001398</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row1_col6\" class=\"data row1 col6\" >0.000183</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row1_col7\" class=\"data row1 col7\" >0.991933</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38level0_row2\" class=\"row_heading level0 row2\" >George W. Bush</th>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row2_col0\" class=\"data row2 col0\" >0.000059</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row2_col1\" class=\"data row2 col1\" >0.000058</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row2_col2\" class=\"data row2 col2\" >0.000058</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row2_col3\" class=\"data row2 col3\" >0.999591</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row2_col4\" class=\"data row2 col4\" >0.000058</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row2_col5\" class=\"data row2 col5\" >0.000059</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row2_col6\" class=\"data row2 col6\" >0.000058</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row2_col7\" class=\"data row2 col7\" >0.000059</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38level0_row3\" class=\"row_heading level0 row3\" >John F. Kennedy</th>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row3_col0\" class=\"data row3 col0\" >0.064080</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row3_col1\" class=\"data row3 col1\" >0.000052</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row3_col2\" class=\"data row3 col2\" >0.000052</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row3_col3\" class=\"data row3 col3\" >0.935608</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row3_col4\" class=\"data row3 col4\" >0.000052</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row3_col5\" class=\"data row3 col5\" >0.000052</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row3_col6\" class=\"data row3 col6\" >0.000052</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row3_col7\" class=\"data row3 col7\" >0.000052</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38level0_row4\" class=\"row_heading level0 row4\" >Lyndon Baines Johnson</th>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row4_col0\" class=\"data row4 col0\" >0.041676</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row4_col1\" class=\"data row4 col1\" >0.000072</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row4_col2\" class=\"data row4 col2\" >0.000072</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row4_col3\" class=\"data row4 col3\" >0.948069</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row4_col4\" class=\"data row4 col4\" >0.000072</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row4_col5\" class=\"data row4 col5\" >0.009895</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row4_col6\" class=\"data row4 col6\" >0.000072</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row4_col7\" class=\"data row4 col7\" >0.000072</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38level0_row5\" class=\"row_heading level0 row5\" >Calvin Coolidge</th>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row5_col0\" class=\"data row5 col0\" >0.540298</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row5_col1\" class=\"data row5 col1\" >0.000038</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row5_col2\" class=\"data row5 col2\" >0.000038</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row5_col3\" class=\"data row5 col3\" >0.187164</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row5_col4\" class=\"data row5 col4\" >0.000038</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row5_col5\" class=\"data row5 col5\" >0.272346</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row5_col6\" class=\"data row5 col6\" >0.000038</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row5_col7\" class=\"data row5 col7\" >0.000039</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38level0_row6\" class=\"row_heading level0 row6\" >Abraham Lincoln</th>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row6_col0\" class=\"data row6 col0\" >0.744337</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row6_col1\" class=\"data row6 col1\" >0.000052</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row6_col2\" class=\"data row6 col2\" >0.000052</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row6_col3\" class=\"data row6 col3\" >0.002066</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row6_col4\" class=\"data row6 col4\" >0.000052</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row6_col5\" class=\"data row6 col5\" >0.000052</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row6_col6\" class=\"data row6 col6\" >0.000052</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row6_col7\" class=\"data row6 col7\" >0.253338</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38level0_row7\" class=\"row_heading level0 row7\" >Donald Trump</th>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row7_col0\" class=\"data row7 col0\" >0.000057</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row7_col1\" class=\"data row7 col1\" >0.000057</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row7_col2\" class=\"data row7 col2\" >0.000057</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row7_col3\" class=\"data row7 col3\" >0.999604</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row7_col4\" class=\"data row7 col4\" >0.000057</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row7_col5\" class=\"data row7 col5\" >0.000057</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row7_col6\" class=\"data row7 col6\" >0.000057</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row7_col7\" class=\"data row7 col7\" >0.000057</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38level0_row8\" class=\"row_heading level0 row8\" >John Adams</th>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row8_col0\" class=\"data row8 col0\" >0.000198</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row8_col1\" class=\"data row8 col1\" >0.000198</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row8_col2\" class=\"data row8 col2\" >0.000198</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row8_col3\" class=\"data row8 col3\" >0.000198</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row8_col4\" class=\"data row8 col4\" >0.000198</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row8_col5\" class=\"data row8 col5\" >0.000198</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row8_col6\" class=\"data row8 col6\" >0.000198</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row8_col7\" class=\"data row8 col7\" >0.998614</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38level0_row9\" class=\"row_heading level0 row9\" >James K. Polk</th>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row9_col0\" class=\"data row9 col0\" >0.318432</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row9_col1\" class=\"data row9 col1\" >0.000020</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row9_col2\" class=\"data row9 col2\" >0.000020</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row9_col3\" class=\"data row9 col3\" >0.000020</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row9_col4\" class=\"data row9 col4\" >0.000020</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row9_col5\" class=\"data row9 col5\" >0.000020</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row9_col6\" class=\"data row9 col6\" >0.000020</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row9_col7\" class=\"data row9 col7\" >0.681450</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38level0_row10\" class=\"row_heading level0 row10\" >James Monroe</th>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row10_col0\" class=\"data row10 col0\" >0.006500</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row10_col1\" class=\"data row10 col1\" >0.000074</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row10_col2\" class=\"data row10 col2\" >0.000074</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row10_col3\" class=\"data row10 col3\" >0.000074</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row10_col4\" class=\"data row10 col4\" >0.000074</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row10_col5\" class=\"data row10 col5\" >0.000074</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row10_col6\" class=\"data row10 col6\" >0.000074</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row10_col7\" class=\"data row10 col7\" >0.993056</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38level0_row11\" class=\"row_heading level0 row11\" >Herbert Hoover</th>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row11_col0\" class=\"data row11 col0\" >0.576017</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row11_col1\" class=\"data row11 col1\" >0.000056</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row11_col2\" class=\"data row11 col2\" >0.000056</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row11_col3\" class=\"data row11 col3\" >0.344884</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row11_col4\" class=\"data row11 col4\" >0.000056</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row11_col5\" class=\"data row11 col5\" >0.078820</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row11_col6\" class=\"data row11 col6\" >0.000056</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row11_col7\" class=\"data row11 col7\" >0.000056</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38level0_row12\" class=\"row_heading level0 row12\" >Theodore Roosevelt</th>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row12_col0\" class=\"data row12 col0\" >0.088512</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row12_col1\" class=\"data row12 col1\" >0.000019</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row12_col2\" class=\"data row12 col2\" >0.000019</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row12_col3\" class=\"data row12 col3\" >0.000020</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row12_col4\" class=\"data row12 col4\" >0.000019</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row12_col5\" class=\"data row12 col5\" >0.908835</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row12_col6\" class=\"data row12 col6\" >0.000019</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row12_col7\" class=\"data row12 col7\" >0.002558</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38level0_row13\" class=\"row_heading level0 row13\" >Dwight D. Eisenhower</th>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row13_col0\" class=\"data row13 col0\" >0.122079</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row13_col1\" class=\"data row13 col1\" >0.000050</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row13_col2\" class=\"data row13 col2\" >0.000050</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row13_col3\" class=\"data row13 col3\" >0.814081</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row13_col4\" class=\"data row13 col4\" >0.000050</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row13_col5\" class=\"data row13 col5\" >0.053987</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row13_col6\" class=\"data row13 col6\" >0.000050</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row13_col7\" class=\"data row13 col7\" >0.009656</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38level0_row14\" class=\"row_heading level0 row14\" >George Herbert Walker Bush</th>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row14_col0\" class=\"data row14 col0\" >0.000074</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row14_col1\" class=\"data row14 col1\" >0.000074</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row14_col2\" class=\"data row14 col2\" >0.000074</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row14_col3\" class=\"data row14 col3\" >0.999483</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row14_col4\" class=\"data row14 col4\" >0.000074</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row14_col5\" class=\"data row14 col5\" >0.000074</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row14_col6\" class=\"data row14 col6\" >0.000074</td>\n",
       "                        <td id=\"T_e2291480_6319_11eb_97c0_39a9936bdd38row14_col7\" class=\"data row14 col7\" >0.000074</td>\n",
       "            </tr>\n",
       "    </tbody></table>"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x7f6be4c60450>"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "president_topic.sample(15).style.background_gradient(cmap=cm)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "#doc_topic.style.background_gradient(cmap=cm)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Topic 7 seems to be related to economic programs, let's plot the prevalence of that topic over time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x7f6be4b53c90>"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 648x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.relplot(x=\"year\",y=7,data=doc_topic,kind=\"line\",height=6,aspect=1.5)\n",
    "#plt.xlim(1900,2000)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Predict Year of Address"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LassoCV\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import mean_absolute_error\n",
    "\n",
    "X = doc_topic.drop(['MostCommonTopic','president','year'],axis=1)\n",
    "y = doc_topic['year']\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X,y, test_size=0.3, random_state=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9622668917308718\n",
      "0.9526408586778637\n"
     ]
    }
   ],
   "source": [
    "LASSOreg = LassoCV(cv=5, random_state=0).fit(X_train, y_train)\n",
    "\n",
    "# Print train and test R^2\n",
    "print(LASSOreg.score(X_train,y_train))\n",
    "print(LASSOreg.score(X_test,y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "11.25343909476925"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mean_absolute_error(LASSOreg.predict(X_test),y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "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>coefficient</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>96.730997</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>5</td>\n",
       "      <td>15.981584</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>-0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>-0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>-0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>-0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>6</td>\n",
       "      <td>-0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>7</td>\n",
       "      <td>-77.997459</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   term  coefficient\n",
       "3     3    96.730997\n",
       "5     5    15.981584\n",
       "0     0    -0.000000\n",
       "1     1    -0.000000\n",
       "2     2    -0.000000\n",
       "4     4    -0.000000\n",
       "6     6    -0.000000\n",
       "7     7   -77.997459"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Obtain pandas dataframe of model coefficients sorted by value\n",
    "# To do : sort by absolute value\n",
    "def obtain_coefficients(model):\n",
    "    coefficients = pd.DataFrame(model.coef_.T)\n",
    "    coefficients.insert(0,'',list(X.columns.values))\n",
    "    coefficients.columns = ['term','coefficient']\n",
    "    coefficients.sort_values(by='coefficient',ascending=False,inplace=True)\n",
    "    return(coefficients)\n",
    "\n",
    "obtain_coefficients(LASSOreg)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "According to the model, topic 4 is associated with more recent addresses while topics 1 and 0 are associated with older addresses. \n",
    "Topic 4 has personal terms such as 'child','family', and 'job' while topics 0 and 1 seem to be more associated with legislation and policy."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Sentiment Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "sentiments = [TextBlob(text).sentiment for text in speech_list ] \n",
    "\n",
    "speeches_pd['polarity'] = [ i[0] for i in sentiments]\n",
    "speeches_pd['subjectivity'] = [ i[1] for i in sentiments]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x7f6be44df050>"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 648x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.relplot(x=\"year\", y=\"polarity\", data=speeches_pd,kind=\"line\",height=6, aspect=1.5, marker=\"o\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}