hist3368-week10-optional-using-wordnet-to-study-congress.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Hist 3368 - Using Wordnet to Study Congress\n",
    "\n",
    "#### By Jo Guldi"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The notebook below generates some visualizations of a controlled vocabulary of crime, applying the findings from Wordnet to Jane Austen and Congress."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import nltk, numpy, re, matplotlib\n",
    "import pandas as pd\n",
    "import re\n",
    "from textblob import TextBlob\n",
    "from nltk.corpus import wordnet as wn\n",
    "from nltk.corpus import stopwords\n",
    "stopwords = stopwords.words('english')\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Load the Controlled vocabulary on crime that we generated earlier this week."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/users/jguldi/digital-history\n"
     ]
    }
   ],
   "source": [
    "cd ~/digital-history"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "You should have a file called \"crime_vocab.csv\" that we created in an earlier notebook using Wordnet.  Load it now."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "vocab = pd.read_csv(\"crime_vocab.csv\").drop(['Unnamed: 0'], 1)\n",
    "vocab = list(vocab['word'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This may take a minute."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Load Austen Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/scratch/group/history/hist_3368-jguldi\n"
     ]
    }
   ],
   "source": [
    "cd /scratch/group/history/hist_3368-jguldi"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open('senseandsensibility.txt', 'r') as myfile:\n",
    "    sas_data = myfile.read().split('\\n\\n\"I suppose you know, ma\\'am, that Mr. Ferrars is married\"\\n\\nIt _was_ Edward\\n\\n\"Everything in such respectable condition\"\\n\\n ')[1].split('THE END')[0].strip()\n",
    "with open('emma.txt', 'r') as myfile:\n",
    "    emma_data = myfile.read().split('CHAPTER I')[1].split('FINIS')[0].strip()\n",
    "with open('prideandprejudice.txt', 'r') as myfile:\n",
    "    pap_data = myfile.read().split('CHAPTER I')[1].split('End of the Project Gutenberg EBook of Pride and Prejudice, by Jane Austen')[0].strip()\n",
    "data = [sas_data, emma_data, pap_data]\n",
    "\n",
    "# clean up\n",
    "austenwords = []\n",
    "for novel in data:\n",
    "    novel = re.sub('[\\\",.;:?([)\\]_*]', '', novel) # strip punctuation\n",
    "    novel = re.sub('-', ' ', novel) # replace hyphens with spaces\n",
    "    novelwords = novel.replace('\\n', ' ').lower().split() # remove whitespace, lowercase, split into words\n",
    "    austenwords += novelwords"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Find Vocab Matches in Jane Austen"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "off        62\n",
       "attempt    39\n",
       "remove     31\n",
       "wrong      24\n",
       "liberty    19\n",
       "dtype: int64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# search for matched words\n",
    "matchedwords = []\n",
    "for w in austenwords:\n",
    "    w = wn.morphy(w) # lemmatize it\n",
    "    if w in vocab:\n",
    "        matchedwords.append(w)\n",
    "matchedwords\n",
    "pd.Series.value_counts(matchedwords)[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['criminal offense',\n",
       " 'criminal offence',\n",
       " 'law breaking',\n",
       " 'aggravated assault',\n",
       " 'assault and battery']"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# collect the bigrams in the controlled vocabulary\n",
    "cleanwordnetbigrams = []\n",
    "for v in vocab:\n",
    "    if \" \" in v:\n",
    "        cleanwordnetbigrams.append(v)\n",
    "cleanwordnetbigrams[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "break in    2\n",
       "take in     2\n",
       "dtype: int64"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# search for matched bigrams\n",
    "matchedbigrams = []\n",
    "from textblob import TextBlob\n",
    "for novel in data:\n",
    "    novel = re.sub('[\\\",.;:?([)\\]_*]', '', novel) # strip punctuation\n",
    "    novel = re.sub('-', ' ', novel) # replace hyphens with spaces\n",
    "    novelwords = novel.replace('\\n', ' ').lower() # remove whitespace, lowercase, split into words\n",
    "    bigrams = TextBlob(novelwords).ngrams(n=2)\n",
    "    for bigram in bigrams: # move through each line of the *bigrams* list\n",
    "        if wn.morphy(bigram[0]) is not None and wn.morphy(bigram[1]) is not None:\n",
    "            bigram2 = wn.morphy(bigram[0]) + ' ' + wn.morphy(bigram[1]) # call the first word, a space, and the second word into a new string\n",
    "            if bigram2 in cleanwordnetbigrams:\n",
    "                matchedbigrams.append(bigram2) # save the string \n",
    "pd.Series.value_counts(matchedbigrams)[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "evil       17\n",
       "offence    14\n",
       "torture     6\n",
       "attack      6\n",
       "injury      5\n",
       "dtype: int64"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# put it all together \n",
    "allcontrolledvocab = matchedwords + matchedbigrams  \n",
    "\n",
    "# stopword out ambiguous words\n",
    "ambiguouswords = ['break in', 'off','setup,' 'foul', 'wrong', 'liberty', 'attempt', 'liquidation', 'petty', 'battery', 'gaming', 'intrusion','raid', 'cheat', 'trick','deception','skimming', 'distort','bluff', 'distortion', 'misrepresentation','execute','gore', 'job', 'technical', 'damage', 'elimination', 'take in', 'burke', 'remove', 'fall', 'familiarity', 'pretence', 'maintenance', 'spoil', 'commission', 'protection', 'execution', 'tribute', 'hit', 'dispatch', 'play', 'rolling']\n",
    "finallist = []\n",
    "for v in allcontrolledvocab:\n",
    "    if v not in ambiguouswords:\n",
    "        finallist.append(v)\n",
    "finalcount = pd.Series.value_counts(finallist)\n",
    "finalcount[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAp0AAAKXCAYAAADATvFYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi41LCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvSM8oowAAIABJREFUeJzs3XucXWV97/HPl4jhHkWQBqpEMIoIEiWgXAWlasW7KCoq4CXVtiK1VGOlFG09DepRaq3alCpW0YJ4VCSoeCGEixAmEBJQ1COEo4AXRMJNAcPv/LHXyGacZGaSWdl7Zj7v12tee+1nPetZv7UzL/jmWc/aSVUhSZIktWmTXhcgSZKkyc/QKUmSpNYZOiVJktQ6Q6ckSZJaZ+iUJElS6wydkiRJap2hU9JGl+TkJJ9bj+OemOSqJHcmOa6N2vpZklVJDhtFv4OS/LClGr6e5Og2xpY0uRk6pSkuybuTnDek7cdraXvVxq3uj7wTWFxVW1fVRzdkoCQzk1SSHbra3rOWtm9syLk2tqq6qKqe2NLYf15Vn1mfY5MsTvKm8a5pjOf/TZLp4zTe6Un+eTzGkqYCQ6ekJcABSaYBJPkTYFPgaUPaHt/0HbV0jOd/Z3YGrl2fA5M8rPt9Vd0C/F/g4K7mg4Hrhmkb9roHPx/1vySzgIOAAl7U02KkKcrQKekKOiFzTvP+YOAC4IdD2n5SVTcDJNk/yRVJVjev+w8O1swmvT/JJcA9wC5JHpfkwua2+LeA7br6b5bkc0l+neT2Zrw/zDR29fsucCjwsSR3JXlCkhlJ/jvJr5LcmOTEwZCb5JgklyT5SJLbgJOHufYlzbUNBsinAv86pG2/pt/gzNYnkpyX5G7g0FHUcHGSDzUzbDck+fOua3pckiXN5/LtJP/evewgyeuaMX+d5D1DPo/pSU5NcnPzc+rgDF6SQ5L8rKvvqiQnJFnR/JmdmWSzrv3vTHJLM86bmtnexw/zeT1ktnKk61uXJI9Mcm7zuf2m2f7TIef5p+bP8M4k5yfp/r15RpJLm9+Zq5McMsIpXw9cBpwOPGR5QIbMwA5eV7Od5nfol81ntyLJHknmAUcB72x+H7/W9N8xyZea67ohXctA0llWclbz+3JnkmuTzB3N5yVNBoZOaYqrqvuAy3lwdu9g4CLg4iFtg8FrW2AR8FHgUcCHgUVJHtU17OuAecDWwI3A54FldMLmP/HQ/+kfDcwAHtOM9xbgt8PU+aymrr+uqq2q6kfAvzXH7gI8k06wOLbrsKcD1wOPBt4/zOX/IXTSCZzXAd8Z0rYpsLTrmNc0Y23dfEajqeGHzbV/APivJGn2fb4Z+1F0QvHrBg9KsjvwiaZtx6bPH0IZ8B7gGXT+YrAXsC9w4jDXOOiVwPOAxwFPAY5pzvM84B3AYXRms5+5jjGGs67rW5dNgE/Tmb1+LJ0/848N6fMaOp/lo4GHAyc0Ne9E53fwn4Ftm/YvJdl+Hed7PXBG8/PcDPMXm7V4Dp3fhycAjwCOBH5dVQubsT7Q/D6+sPnLxteAq4GdgGcDxyd5btd4LwL+pxnrnGGuWZq0DJ2SAC7kwaB1EJ1wd9GQtgub7cOBH1fVZ6vq91X1BTph7YVd451eVddW1e+BmcA+wD9U1b1VtYTO/5gH3U8nUD2+qtZU1bKqumOkgptZyCOBd1fVnVW1CvjfdAU34Oaq+remzj8Kss017ZHkkYPXXVU/BrbrarusCeaDvlpVl1TVA03tI9VwY1X9Z1WtAT7TfB47JHls87mcVFX3VdXFdELIoCOAc6tqSVXdC/wD8EDX/qOA91XVL6vqV8B7h5x3qI9W1c1VdRudz39wFvuVwKebP697mnHGYtjrG+mgqvp1VX2pqu6pqjvpBPmhgffTVfWj5s/urK6aXwucV1XnVdUDVfUtYAB4/nDnSnIgnXB7VlUtA35CJ9COxv10/oKxG5Cq+kGzNGM4+wDbV9X7mj/T64H/BLrXQl/c1L0G+CydvzBIU4KhUxJ0ZvwObILW9k3wuhTYv2nbgwfXNe5IZ/ay2410ZnYG/bRre0fgN1V195D+gz4LfBP4n+b27geSbDqKmrejM/vVPda66vgjTUj8GXAgD87wAnyvq23oes7uMUdTw8+7zndPs7kVnc/ltq62oWPv2P2++fx+PWT/0PPuyNr9vGv7nqaGPzoPI3xm6xp3yPWtU5ItkvxHs3zgDjqf8yPy0HWya6t5Z+AVza3125PcTufPa+ZaTnc0cH5V3dq8/zxDbrGvTVV9l85s5L8Dv0iyMMk2a+m+M7DjkLr+noeG8KHXtFmGrDeWJitDpyTohKwZdG6JXwLQzDbe3LTdXFU3NH1vpvM/126PBW7qel9d27cAj0yy5ZD+NOe5v6reW1W7A/sDL6BzK3Qkt9KZhequZV11rM3gjO5+dIJ2d9uB/HHo7B5zNDWszS3Atkm26Gp7zJD9f3jf9OtewjD0z+GxTdtY3cJDb9s/Zm0dx9nfAk8Enl5V2/DgrPpobs3/FPhsVT2i62fLqlowtGOSzenM5j4zyc+T/Bz4G2CvJIOzjHcD3X8Of9I9RlV9tKr2Bp5M5zb73w3uGqauG4bUtXVVDTsDK001hk5JNLcvB+is7buoa9fFTVt38DoPeEKS1yR5WJIjgd2Bc9cy9o3N2O9N8vDmVucfbsUnOTTJns0M1x10QtyaUdS8hs4t1/cn2TrJzk2tY/3+zyV0Qu7NXbf1L27aZtAJ5ONeQ9fncnLzuezHQ5conA28IMmBSR4OvI+H/jf7C8CJSbZvHrA5aTTnHcZZwLFJntQE25PWY4z1sTWddZy3N+uE/3EMx34OeGGS5yaZls7DaId0P4jU5SV0fp92p3N7fg7wJDq/54N/uVkOvKyZfX088MbBg5Psk+Tpzez73cDvePD38xd01vIOWgrckeRdSTZvatsjyT5juDZp0jJ0Shp0IZ0HNi7uaruoaftD6KyqX9OZjfxbOrd73wm8oOvW5XBeQ+eBk9vohIv/7tr3J3QC1h3AD5o6Rhue3kYnCFzf1P154FOjPHbQcNe9HNgcWDbk9vd413AUnRnWX9N5KOZM4F6AqroW+KtmvFuA39BZCjDon+mE1hXASuDKpm1MqurrdB4Ku4DOV0gNhux7xzrWaE/ZvJ5K5zO+lc5T5aP+LtSq+inwYjq3rn9FZ4bx7xj+/2lH01kb+v+q6ueDP3RumR/V3Nr+CHAfnRD5GToPCA3ahs66zN/QWcLwa+BDzb7/AnZvbqV/pflLyAvpBNsbmms7jc5fXqQpL1WjufskSWpbkjOB66pqLLN+413Dk4BrgOnNg2DjOfaVdB5++sp4jitpYnCmU5J6pLl1u2uSTZqvLnoxsNEDWZKXNrf4HwmcAnythcD5ZDq3ta8az3ElTRyGTknqnT8BFgN30bnF/daq6kUo+ws6t6l/Qme94lvHc/AkpwDnA+9q1rJKmoK8vS5JkqTWOdMpSZKk1hk6JUmS1DpDpyRJklpn6JQkSVLrDJ2SJElqnaFTkiRJrTN0SpIkqXWGTkmSJLXO0ClJkqTWGTolSZLUOkOnJEmSWmfolCRJUusMnZIkSWqdoVOSJEmtM3RKkiSpdYZOSZIktc7QKUmSpNYZOiVJktQ6Q6ckSZJaZ+iUJElS6wydkiRJap2hU5IkSa0zdEqSJKl1hk5JkiS1ztApSZKk1hk6JUmS1DpDpyRJklpn6JQkSVLrDJ2SJElqnaFTkiRJrTN0SpIkqXWGTkmSJLXO0ClJkqTWGTolSZLUOkOnJEmSWmfolCRJUusMnZIkSWqdoVOSJEmtM3RKkiSpdYZOSZIktc7QKUmSpNYZOiVJktQ6Q6ckSZJaZ+iUJElS6wydkiRJap2hU5IkSa0zdEqSJKl1hk5JkiS1ztApSZKk1hk6JUmS1DpDpyRJklpn6JQkSVLrDJ2SJElqnaFTkiRJrTN0SpIkqXWGTkmSJLXO0ClJkqTWPazXBeihtttuu5o1a1avy5AkSRrRsmXLbq2q7UfT19DZZ2bNmsXAwECvy5AkSRpRkhtH29fb65IkSWqdoVOSJEmtM3RKkiSpdYZOSZIktc7QKUmSpNYZOiVJktQ6Q6ckSZJaZ+iUJElS6wydkiRJap2hU5IkSa0zdEqSJKl1hk5JkiS1ztApSZKk1hk6JUmS1DpDpyRJklpn6JQkSVLrDJ2SJElqnaFTkiRJrTN0SpIkqXWGTkmSJLXO0ClJkqTWGTolSZLUOkOnJEmSWvewXhegh1p502pmzV/U6zIkSdIEtmrB4b0u4Y840ylJkqTWGTolSZLUOkOnJEmSWmfolCRJUusMnZIkSWqdobNlSS5tXmcluabX9UiSJPWCobNlVbV/r2uQJEnqNb+ncz0keS1wHPBw4HJgBbBzVb2z2X8MsHdVvS3JXVW1Vc+KlSRJ6gPOdI5RkicBRwIHVNUcYA1wF/Cyrm5HAmeOYcx5SQaSDKy5Z/W41itJktQPnOkcu2cDewNXJAHYHPglcH2SZwA/Bp4IXDLaAatqIbAQYPrM2TXeBUuSJPWaoXPsAnymqt79kMbkjcArgeuAL1eV4VGSJKnh7fWx+w5wRJJHAyTZNsnOwP8BXgK8mjHcWpckSZoKDJ1jVFXfB04Ezk+yAvgWMLOqfgN8n84DRUt7WaMkSVK/8fb6eqiqMxlmNrOqXjBM21bN6ypgj9aLkyRJ6kPOdEqSJKl1hk5JkiS1ztApSZKk1rmms8/sudMMBhYc3usyJEmSxpUznZIkSWqdoVOSJEmtM3RKkiSpdYZOSZIktc4HifrMyptWM2v+ol6XIUnSuFjlw7FqONMpSZKk1hk6JUmS1DpDpyRJklpn6JQkSVLrDJ2SJElq3ZQNnUmOS/KDJGckmZ7k20mWJzmy17VJkiRNNlP5K5P+EvjzqrohyTOATatqTq+LkiRJmoymxExnknckuab5OT7JJ4FdgHOSvAv4HDCnmencNcneSS5MsizJN5PMbMZZnOSUJEuT/CjJQU37tCQfSrIyyYokb2vahx1HkiRpqpn0M51J9gaOBZ4OBLgceC3wPODQqro1yeXACVX1giSbAp8FXlxVv2put78feEMz5MOqat8kzwf+ETgMmAc8DnhqVf0+ybbNOP+2jnG6a5zXjMG0bbZv6ZOQJEnqnUkfOoEDgS9X1d0ASf4PcNA6+j8R2AP4VhKAacAtXfv/T/O6DJjVbB8GfLKqfg9QVbcl2WOEcf6gqhYCCwGmz5xdY7s8SZKk/jcVQmfWo/+1VbXfWvbf27yu4cHPL8DQsDjSOJIkSVPGVFjTuQR4SZItkmwJvBS4aB39fwhsn2Q/gCSbJnnyCOc4H3hLkoc1x2y7nuNIkiRNSpM+dFbVlcDpwFI66zlPq6qr1tH/PuAI4JQkVwPLgf1HOM1pwP8DVjTHvGY9x5EkSZqUUuUSwn4yfebsmnn0qb0uQ5KkcbFqweG9LkEtSrKsquaOpu+kn+mUJElS7xk6JUmS1DpDpyRJklo3Fb4yaULZc6cZDLj+RZIkTTLOdEqSJKl1hk5JkiS1ztApSZKk1rmms8+svGk1s+Yv6nUZ0kbh9/dJ0tThTKckSZJaZ+iUJElS6wydkiRJap2hU5IkSa0zdEqSJKl1Uyp0JnlEkr9s+xhJkiQ91JQKncAjgFEHyCTTxnrMkGMlSZLE1AudC4BdkyxP8sHm55okK5McCZDkkCQXJPk8sHKYYw5Jcu7ggEk+luSYZntVkpOSXAy8IsmuSb6RZFmSi5LstvEvWZIkqfem2pfDzwf2qKo5SV4OvAXYC9gOuCLJkqbfvk2/G5LMGjwGOqF0hHP8rqoObPp+B3hLVf04ydOBjwPPGudrkiRJ6ntTLXR2OxD4QlWtAX6R5EJgH+AOYGlV3bCe454JkGQrYH/gi0kG900f7oAk84B5ANO22X49TytJktS/pnLozDr23b2Ofb/nocsSNlvLsZsAtw/OkK5LVS0EFgJMnzm7RuovSZI00Uy1NZ13Als320uAI5NMS7I9cDCwdIRjAG4Edk8yPckM4NnDnaiq7gBuSPIKgHTsNU7XIUmSNKFMqdBZVb8GLklyDbAfsAK4Gvgu8M6q+vm6jknywar6KXBWc+wZwFXrOOVRwBuTXA1cC7x4XC9IkiRpgkiVd3P7yfSZs2vm0af2ugxpo1i14PBelyBJ2gBJllXV3NH0nVIznZIkSeoNQ6ckSZJaZ+iUJElS6wydkiRJat1U/p7OvrTnTjMY8OEKSZI0yTjTKUmSpNYZOiVJktQ6Q6ckSZJa55rOPrPyptXMmr+o12VIG4VfDi9JU4cznZIkSWqdoVOSJEmtM3RKkiSpdYZOSZIktc7QKUmSpNYZOoeR5O+7th+R5C83YKzTkxwxPpVJkiRNTIbO4f191/YjgPUOnZIkSfJ7OknyFeAxwGbAvwK7AJsnWQ5cC0wDdm3efwt4L/BV4JHApsCJVfXVZqzXAycABayoqtcNOdc/Ned6Q1U9sBEuT5IkqS9M+dBJJwDelmRz4ArgmcBfV9UcgCSzgD263j8MeGlV3ZFkO+CyJOcAuwPvAQ6oqluTbNt9kiQfAGYAx1ZVDdk3D5gHMG2b7du7UkmSpB4xdMJxSV7abD8GmD1C/wD/K8nBwAPATsAOwLOAs6vqVoCquq3rmH8ALq+qecMNWFULgYUA02fOruH6SJIkTWRTOnQmOQQ4DNivqu5JspjObfZ1OQrYHti7qu5Psqo5JnRuqw/nCmDvJNsOCaOSJElTwlR/kGgG8JsmcO4GPKNpvz/Jps32ncDWQ475ZRM4DwV2btq/A7wyyaMAhtxe/wawAFiUpHssSZKkKWFKz3TSCYNvSbIC+CFwWdO+EFiR5MqqOirJJUmuAb4OnAJ8LckAsBy4DqCqrk3yfuDCJGuAq4BjBk9UVV9sAuc5SZ5fVb/dSNcoSZLUcxnyTIt6bPrM2TXz6FN7XYa0UaxacHivS5AkbYAky6pq7mj6TvXb65IkSdoIDJ2SJElqnaFTkiRJrZvqDxL1nT13msGA69wkSdIk40ynJEmSWmfolCRJUusMnZIkSWqdoVOSJEmt80GiPrPyptXMmr+o12VoCvEL2iVJG4MznZIkSWqdoVOSJEmtM3RKkiSpdYZOSZIktc7QKUmSpNYZOhtJLh1Fn9OS7L4x6pEkSZpM/MqkRlXtP4o+bxrLmEkeVlW/X/+qJEmSJgdnOhtJ7mpeD0myOMnZSa5LckaSNPsWJ5nb3b/ZPiLJ6c326Uk+nOQC4INJfpxk+2bfJkn+b5LtNvb1SZIk9ZIzncN7KvBk4GbgEuAA4OIxHP8E4LCqWpPkduAo4FTgMODqqrq1u3OSecA8gGnbbL/h1UuSJPUZZzqHt7SqflZVDwDLgVljPP6LVbWm2f4U8Ppm+w3Ap4d2rqqFVTW3quZO22LG+tYsSZLUtwydw7u3a3sNw88IV9f2ZkP23f2HTlU/BX6R5FnA04Gvj1eRkiRJE4Whc/39IsmTkmwCvHSEvqcBnwPO6poBlSRJmjIMnWM3OMM5HzgX+C5wywjHnANsxTC31iVJkqYCHyRqVNVWzetiYHFX+193dXsUcFvTfjZw9jDjHDPM8HvReYDounErWJIkaQIxdI5Skm8BK6vqhjEeNx94K50n2CVJkqYkQ+coVdWfredxC4AF41yOJEnShOKaTkmSJLXOmc4+s+dOMxhYcHivy5AkSRpXznRKkiSpdYZOSZIktc7QKUmSpNYZOiVJktQ6HyTqMytvWs2s+Yt6XYamkFU+uCZJ2gic6ZQkSVLrDJ2SJElqnaFTkiRJrTN0SpIkqXWTLnQmuXQUfU5Lsvt6jD0nyfO73r8oyfyxjiNJkjTVTLqn16tq/1H0edN6Dj8HmAuc14xzDnDOeo4lSZI0ZUzGmc67mtdDkixOcnaS65KckSTNvsVJ5nb3b7aPSHJ6s/2KJNckuTrJkiQPB94HHJlkeZIjkxyT5GNN/x2SfLnpf3WS/Zv21yZZ2hzzH0mmbdQPRJIkqQ9MutA5xFOB44HdgV2AA8Zw7EnAc6tqL+BFVXVf03ZmVc2pqjOH9P8ocGHT/2nAtUmeBBwJHFBVc4A1wFEbdEWSJEkT0GQPnUur6mdV9QCwHJg1hmMvAU5P8mZgNLOTzwI+AVBVa6pqNfBsYG/giiTLm/e7DD0wybwkA0kG1tyzegwlSpIkTQyTbk3nEPd2ba9h+Outru3N/tBY9ZYkTwcOB5YnmbMe5w/wmap697o6VdVCYCHA9Jmza119JUmSJqLJPtM5Gr9I8qQkmwAvHWxMsmtVXV5VJwG3Ao8B7gS2Xss43wHe2hw7Lck2TdsRSR7dtG+bZOcWr0WSJKkvTeXQOTijOB84F/gucEvX/g8mWZnkGmAJcDVwAbD74INEQ8Z7O3BokpXAMuDJVfV94ETg/CQrgG8BM1u7IkmSpD6Vqql3N7cJhi+qqht6XctQ02fOrplHn9rrMjSFrFpweK9LkCRNUEmWVdXc0fSdcjOdSb4FrOzHwClJkjRZTfYHif5IVf1Zr2uQJEmaaqbcTKckSZI2PkOnJEmSWjflbq/3uz13msGAD3ZIkqRJxplOSZIktc7QKUmSpNYZOiVJktQ613T2mZU3rWbW/EW9LkNTiF8OL0naGJzplCRJUusMnZIkSWqdoVOSJEmtM3RKkiSpdYZOSZIktW7Chs4kxyX5QZIz1rL/kCTnrse4s5Jcs+EVbpxxJUmSJoKJ/JVJfwn8eVXd0OtCJEmStG4TcqYzySeBXYBzkrwryaVJrmpenzhM/2cmWd78XJVk66b975JckWRFkvd2HTItyX8muTbJ+Uk2b/rPSXJZ0//LSR45QvveSa5O8j3gr9r+XCRJkvrVhAydVfUW4GbgUOATwMFV9VTgJOB/DXPICcBfVdUc4CDgt0meA8wG9gXmAHsnObjpPxv496p6MnA78PKm/b+Bd1XVU4CVwD+O0P5p4Liq2m9d15NkXpKBJANr7lk9lo9CkiRpQpiQoXOIGcAXm/WSHwGePEyfS4APJzkOeERV/R54TvNzFXAlsBudsAlwQ1Utb7aXAbOSzGiOvbBp/wxw8BjaP7u2C6iqhVU1t6rmTttixpg/AEmSpH43GULnPwEXVNUewAuBzYZ2qKoFwJuAzYHLkuwGBPiXqprT/Dy+qv6rOeTersPXsH5rXwPUehwnSZI06UyG0DkDuKnZPma4Dkl2raqVVXUKMEBnVvObwBuSbNX02SnJo9d2kqpaDfwmyUFN0+uAC9fRfjuwOsmBTftR632FkiRJE9xEfnp90AeAzyR5B/DdtfQ5PsmhdGYtvw98varuTfIk4HtJAO4CXtv0WZujgU8m2QK4Hjh2hPZjgU8luYdOyJUkSZqSUuUd4H4yfebsmnn0qb0uQ1PIqgWH97oESdIElWRZVc0dTd/JcHtdkiRJfc7QKUmSpNYZOiVJktS6yfAg0aSy504zGHCNnSRJmmSc6ZQkSVLrDJ2SJElqnaFTkiRJrTN0SpIkqXU+SNRnVt60mlnzF/W6DE0hfjm8JGljcKZTkiRJrTN0SpIkqXWGTkmSJLXO0ClJkqTWGTolSZLUur4LnUku7XUNa5PktCS7j9Dn+CRbbKyaJEmSJoK+C51Vtf94j5lkg78aKsm0qnpTVX1/hK7HA4ZOSZKkLn0XOpPc1bwekuTCJGcl+VGSBUmOSrI0ycokuzb9Tk/yySQXNf1e0LQfk+SLSb4GnJ+ODya5pjn+yK7zLEny5STfb8baZLCWJO9LcjmwX5LFSeY2+56T5HtJrmzOs1WS44AdgQuSXJDkjUk+0nVtb07y4Y35eUqSJPWDfv9y+L2AJwG3AdcDp1XVvkneDryNzqwiwCzgmcCudALf45v2/YCnVNVtSV4OzGnG3A64IsmSpt++wO7AjcA3gJcBZwNbAtdU1UkASWhetwNOBA6rqruTvAt4R1W9L8k7gEOr6tYkWwIrkryzqu4HjgX+YuhFJpkHzAOYts32G/qZSZIk9Z2+m+kc4oqquqWq7gV+ApzftK+kEzQHnVVVD1TVj+mE092a9m9V1W3N9oHAF6pqTVX9ArgQ2KfZt7Sqrq+qNcAXmr4Aa4AvDVPXM+iE1EuSLAeOBnYe2qmq7ga+C7wgyW7AplW1cph+C6tqblXNnbbFjJE+E0mSpAmn32c67+3afqDr/QM8tPYactzg+7u72rKO86zt+N81QXSo0Am0r17HmINOA/4euA749Cj6S5IkTTr9PtM5Wq9IskmzznMX4IfD9FkCHJlkWpLtgYOBpc2+fZM8rlnLeSRw8Qjnuww4YPA2fpItkjyh2XcnsPVgx6q6HHgM8Bo6s6iSJElTzmQJnT+kc7v868Bbqup3w/T5MrACuJrOLe93VtXPm33fAxYA1wA3NH3Xqqp+BRwDfCHJCjohdPCW/kLg60ku6DrkLOCSqvrN2C9NkiRp4kvV0DvLE0uS04Fzq+rs9Tz+EOCEqnrBeNY15BznAh+pqu+M1Hf6zNk18+hT2ypF+iOrFhze6xIkSRNUkmVVNXc0fSfLTGdfSvKIJD8CfjuawClJkjRZ9fuDRCOqqmM28PjFwOLxqGWYsW8HnjBiR0mSpEnOmU5JkiS1bsLPdE42e+40gwHX2EmSpEnGmU5JkiS1ztApSZKk1hk6JUmS1DpDpyRJklrng0R9ZuVNq5k1f1Gvy5hy/IJ0SZLa5UynJEmSWmfolCRJUusMnZIkSWqdoVOSJEmtM3SOQZIXJZnf6zokSZImGp9eH6UkD6uqc4Bzel2LJEnSRGPo7JLk9cAJQAErgDXAbcBTgSuTrATmVtVfJzkd+C2wG7AzcCxwNLAfcHlVHdOM+RzgvcB04CfAsVV110a8LEmSpJ7z9nojyZOB9wDPqqq9gLc3u54AHFZVfzvMYY8EngX8DfA14CPAk4E9k8xJsh1wYnP804AB4B3tXokkSVL/cabzQc8Czq6qWwGq6rYkAF+sqjVrOeZrVVXNDOgvqmolQJJrgVnAnwK7A5c0Yz0c+N7QQZLMA+YBTNtm+/G8JkmSpL5g6HxQ6NxWH+rudRxzb/P6QNf24PuH0bk9/62qevW6TlxVC4GFANNnzh7/Ac+/AAAgAElEQVSuBkmSpAnN2+sP+g7wyiSPAkiy7TiMeRlwQJLHN2NukeQJ4zCuJEnShOJMZ6Oqrk3yfuDCJGuAq8ZhzF8lOQb4QpLpTfOJwI82dGxJkqSJJFXeze0n02fOrplHn9rrMqacVQsO73UJkiRNOEmWVdXc0fT19rokSZJaZ+iUJElS6wydkiRJap2hU5IkSa3z6fU+s+dOMxjwoRZJkjTJONMpSZKk1hk6JUmS1DpDpyRJklrnms4+s/Km1cyav6jXZUw5fjm8JEntcqZTkiRJrTN0SpIkqXWGTkmSJLXO0ClJkqTWGTolSZLUOkPnWiS5a4T9c5N8dGPVI0mSNJH5lUnrqaoGgIFe1yFJkjQRONM5giSfTfLirvdnJHlRkkOSnNu0nZzkU0kWJ7k+yXFd/f8hyXVJvpXkC0lO6MV1SJIk9ZKhc2SnAccCJJkB7A+cN0y/3YDnAvsC/5hk0yRzgZcDTwVeBswd7gRJ5iUZSDKw5p7VLVyCJElSbxk6R1BVFwKPT/Jo4NXAl6rq98N0XVRV91bVrcAvgR2AA4GvVtVvq+pO4GtrOcfCqppbVXOnbTGjpSuRJEnqHdd0js5ngaOAVwFvWEufe7u219D5bNNyXZIkSROCM52jczpwPEBVXTuG4y4GXphksyRbAf4D35IkaUpypnMUquoXSX4AfGWMx12R5BzgauBGOk+7u2hTkiRNOYbOtaiqrQa3k2wBzAa+0LV/MbC42T55yLF7dL39UFWd3IyxBPjfrRUtSZLUp7y9PoIkhwHXAf9WVeszS7kwyXLgSjoPIV05rgVKkiRNAM50jqCqvg08dgOOf804liNJkjQhOdMpSZKk1jnT2Wf23GkGAwt8yF2SJE0uznRKkiSpdYZOSZIktc7QKUmSpNYZOiVJktQ6HyTqMytvWs2s+Yt6XUZPrPIBKkmSJi1nOiVJktQ6Q6ckSZJaZ+iUJElS6wydkiRJap2hU5IkSa0zdG6gJLOSXDOkbW6SjzbbxyT5WLN9cpITelGnJElSL/mVSS2oqgFgoNd1SJIk9QtnOsdRkl2SXJXk75Kc2+t6JEmS+oWhc5wkeSLwJeBY4IoxHjsvyUCSgTX3rG6lPkmSpF5a5+31JO9Y1/6q+vD4ljNhbQ98FXh5VV2b5JCxHFxVC4GFANNnzq7xL0+SJKm3RlrTuXXz+kRgH+Cc5v0LgSVtFTUBrQZ+ChwAXNvjWiRJkvrOOkNnVb0XIMn5wNOq6s7m/cnAF1uvbuK4D3gJ8M0kdwE397geSZKkvjLaNZ2PpROsBt0HzBr3aiawqrobeAHwN8CMHpcjSZLUV0b7lUmfBZYm+TJQwEuBz7RW1QRSVauAPZrt2+ksQ4DOGk+q6nTg9Gb75I1dnyRJUj8YVeisqvcn+TpwUNN0bFVd1V5ZkiRJmkxGDJ1JNgFWVNUewJXtlyRJkqTJZsQ1nVX1AHB1ksduhHokSZI0CY12TedM4NokS4G7Bxur6kWtVDWF7bnTDAYWHN7rMiRJksbVaEPne1utQpIkSZPaaB8kujDJDjz4ZPbSqvple2VJkiRpMhnV93QmeSWwFHgF8Erg8iRHtFmYJEmSJo/R3l5/D7DP4Oxmku2BbwNnt1WYJEmSJo/Rhs5NhtxO/zWj/9eMNAYrb1rNrPmLel1GT6zyASpJkiat0YbObyT5JvCF5v2RwHntlCRJkqTJZp2hM8nxwCXA3wMvBA4EAiysqi+3X54kSZImg5FmOv8U+FdgN2AFcCmdEPq9luuSJEnSJLLO0FlVJwAkeTgwF9gfeAPwn0lur6rd2y9RkiRJE91oHwbaHNgGmNH83Axc3lZR6yvJyUlOWI/j3pfksBH6HJNkx673pyUxdEuSJI3CSGs6FwJPBu6kEzIvBT5cVb/ZCLVtNFV10ii6HQNcQydwU1VvarMmSZKkyWSkmc7HAtOBnwM3AT8Dbm+7qLFI8p4kP0zybeCJTduuSb6RZFmSi5LslmRGklVJNmn6bJHkp0k2TXL64JfdJzkpyRVJrkmyMB1H0FlecEaS5Uk2T7I4ydzmmFcnWdkcc0pXbXcleX+Sq5Nc1vyrTpIkSVPOOkNnVT2Pzj99+aGm6W+BK5Kcn6Tn/x57kr2BVwFPBV7Gg/9M50LgbVW1N3AC8PGqWg1cDTyz6fNC4JtVdf+QYT9WVftU1R50lhW8oKrOBgaAo6pqTlX9tquGHYFTgGcBc4B9kryk2b0lcFlV7QUsAd48jpcvSZI0YYy4prM6rqHzvZxfp/P0+q7A21uubTQOAr5cVfdU1R3AOcBmdB54+mKS5cB/ADOb/mfS+Y5R6ITVM4cZ89AklydZSSdIPnmEGvYBFlfVr6rq98AZwMHNvvuAc5vtZcCs4QZIMi/JQJKBNfesHuF0kiRJE89IazqPoxPgDgDu58GvS/oUsLL16kanhrzfBLi9quYM0/cc4F+SbAvsDXy3e2eSzYCPA3Or6qdJTqYTYtcl69h3f1UN1reGtXzeVbWQzuws02fOHno9kiRJE95IM52z6Pz76vtW1S5V9bqq+nhVXV1VD7Rf3oiWAC9t1lhuTeeW+T3ADUleAdCsydwLoKruApbS+e7Rc6tqzZDxBgPmrUm2Ao7o2ncnsPUwNVwOPDPJdkmmAa8GLhyfy5MkSZocRvqezndsrELWR1VdmeRMYDlwI3BRs+so4BNJTgQ2Bf6HznpO6NxS/yJwyDDj3Z7kP+nM4q4CrujafTrwySS/BfbrOuaWJO8GLqAz63leVX11nC5RkiRpUsiDd3/VD6bPnF0zjz6112X0xKoFh/e6BEmSNAZJllXV3NH0He2Xw0uSJEnrzdApSZKk1hk6JUmS1DpDpyRJklq3zqfXtfHtudMMBnygRpIkTTLOdEqSJKl1hk5JkiS1ztApSZKk1rmms8+svGk1s+Yv6tn5/YJ2SZLUBmc6JUmS1DpDpyRJklpn6JQkSVLrDJ2SJElqnaFTkiRJrTN0jiDJW5K8vtk+PckRzfZpSXbvbXWSJEkTg1+ZNIKq+uRa2t+0sWuRJEmaqCb1TGeSLZMsSnJ1kmuSHJlkVZJTkixtfh7f9N05yXeSrGheH9u0n5zkhGHGXpxkbrN9V5L3N+e5LMkOTfuuzfsrkrwvyV0b8/olSZL6xaQOncDzgJuraq+q2gP4RtN+R1XtC3wMOLVp+xjw31X1FOAM4KNjOM+WwGVVtRewBHhz0/6vwL9W1T7AzWs7OMm8JANJBtbcs3oMp5UkSZoYJnvoXAkc1sxsHlRVg4nuC12v+zXb+wGfb7Y/Cxw4hvPcB5zbbC8DZnWN+cVm+/OsRVUtrKq5VTV32hYzxnBaSZKkiWFSr+msqh8l2Rt4PvAvSc4f3NXdbW2Hj+FU91fVYP81TPLPVZIkaawm9Uxnkh2Be6rqc8CHgKc1u47sev1es30p8Kpm+yjg4nEo4TLg5c32q9bVUZIkaTKb7DNyewIfTPIAcD/wVuBsYHqSy+mE7lc3fY8DPpXk74BfAceOw/mPBz6X5G+BRYALNiVJ0pSUB+8KTw1JVgFzq+rWjXCuLYDfVlUleRXw6qp68bqOmT5zds08+tR1dWnVqgWH9+zckiRpYkmyrKrmjqbvZJ/p7LW9gY8lCXA78IYe1yNJktQTUy50VtWsjXiui4C9Ntb5JEmS+tWkfpBIkiRJ/WHKzXT2uz13msGA6yolSdIk40ynJEmSWmfolCRJUusMnZIkSWqdoVOSJEmt80GiPrPyptXMmr+oZ+f3y+ElSVIbnOmUJElS6wydkiRJap2hU5IkSa0zdEqSJKl1hk5JkiS1bkqGziTHJPnYRj7n8Um22JjnlCRJ6hcTJnSmY8LUO4zjAUOnJEmakloNcUlem2RpkuVJ/iPJtCR3JXl/kquTXJZkh6bvDkm+3LRfnWT/JLOS/CDJx4ErgcckeXWSlUmuSXJK17k+kWQgybVJ3tvVvk+SS5sxlybZutm1Y5JvJPlxkg909b8rySlJliX5dpJ9kyxOcn2SFzV9piX5YJIrkqxI8hdN+yFN37OTXJfkjCYsHwfsCFyQ5II2P3NJkqR+1FroTPIk4EjggKqaA6wBjgK2BC6rqr2AJcCbm0M+ClzYtD8NuLZpfyLw31X1VOB+4BTgWcAcYJ8kL2n6vaeq5gJPAZ6Z5ClJHg6cCby9Gfcw4LdN/zlNfXsCRyZ5TNO+JbC4qvYG7gT+Gfgz4KXA+5o+bwRWV9U+wD7Am5M8rtn3VDqzmrsDuzTX/1HgZuDQqjp0mM9qXhOYB9bcs3q0H7EkSdKE0ea/SPRsYG/giiQAmwO/BO4Dzm36LKMT6KATJF8PUFVrgNVJHgncWFWXNX32oRMIfwWQ5AzgYOArwCuTzGuuaSad0FfALVV1RTPuHc1xAN+pqtXN++8DOwM/ber7RnO+lcC9VXV/kpXArKb9OcBTkhzRvJ8BzG6OXVpVP2vGXd4cc/G6PqiqWggsBJg+c3atq68kSdJE1GboDPCZqnr3QxqTE6pqMFitGUUNdw8Z849P1JllPAHYp6p+k+R0YLOm/9pC3L1d29113N9V3wOD/arqgSSDfQK8raq+OaSOQ9YxriRJ0pTV5prO7wBHJHk0QJJtk+w8Qv+3Nn2nJdlmmD6X07l1vl2SacCrgQuBbeiE09XNGtE/b/pfR2ft5j7NuFt3BccN8U3grUk2bcZ9QpItRzjmTmDrEfpIkiRNSq3NwlXV95OcCJzfPHV+P/BX6zjk7cDCJG+kM0P4VuCWIWPekuTdwAV0ZhvPq6qvAiS5is460OuBS5r+9yU5Evi3JJvTWc952Dhc3ml0bptfmc69+l8BL1nnEZ3b519Pcstw6zolSZImszx4J1n9YPrM2TXz6FN7dv5VCw7v2bklSdLEkmRZ8yD3iCby915KkiRpgjB0SpIkqXWGTkmSJLXOr/PpM3vuNIMB11VKkqRJxplOSZIktc7QKUmSpNYZOiVJktQ6Q6ckSZJa54NEfWblTauZNX9Rz87vl8NLkqQ2ONMpSZKk1hk6JUmS1DpDpyRJklpn6JQkSVLrDJ3jLMlpSXbvdR2SJEn9xKfXx1lVvanXNUiSJPUbZzo3QJItkyxKcnWSa5IcmWRxkrnN/ruSvL/Zf1mSHXpdsyRJUi8YOjfM84Cbq2qvqtoD+MaQ/VsCl1XVXsAS4M0bu0BJkqR+YOjcMCuBw5KckuSgqlo9ZP99wLnN9jJg1nCDJJmXZCDJwJp7hg4hSZI08Rk6N0BV/QjYm074/JckJw3pcn9VVbO9hrWsoa2qhVU1t6rmTttiRnsFS5Ik9YgPEm2AJDsCt1XV55LcBRzT45IkSZL6kqFzw+wJfDDJA8D9wFuBD/W2JEmSpP5j6NwAVfVN4JtDmg/p2r9V1/bZwNkbpzJJkqT+4ppOSZIktc7QKUmSpNYZOiVJktQ6Q6ckSZJa54NEfWbPnWYwsODwXpchSZI0rpzplCRJUusMnZIkSWqdoVOSJEmtc01nn1l502pmzV/Us/Ovcj2pJElqgTOdkiRJap2hU5IkSa0zdEqSJKl1hk5JkiS1ztApSZKk1vVt6ExycpITkrwvyWEb+dynJzlimPZDkpzbbL8oyfyNWZckSdJE1fdfmVRVJ/W6huFU1TnAOb2uQ5IkaSLoq5nOJO9J8sMk3wae2LT9YdYxyYIk30+yIsmHmrZXJLkmydVJljRt05J8MMkVTd+/aNoPSXJhkrOS/KgZ76gkS5OsTLJrVzmHJbmo6feCYWo9JsnHumr8aJJLk1zfVe8mST6e5Nok5yY5b7gZVEmSpMmub2Y6k+wNvAp4Kp26rgSWde3fFngpsFtVVZJHNLtOAp5bVTd1tb0RWF1V+ySZDlyS5Pxm317Ak4DbgOuB06pq3yRvB94GHN/0mwU8E9gVuCDJ40e4hJnAgcBudGZAzwZe1oyzJ/Bo4AfAp4a59nnAPIBp22w/wmkkSZImnn6a6TwI+HJV3VNVd/DHt67vAH4HnJbkZcA9TfslwOlJ3gxMa9qeA7w+yXLgcuBRwOxm3xVVdUtV3Qv8BBgMoyvpBMRBZ1XVA1X1YzrhdLcR6v9K0//7wA5N24HAF5v2nwMXDHdgVS2sqrlVNXfaFjNGOI0kSdLE00+hE6DWuqPq98C+wJeAlwDfaNrfApwIPAZYnuRRQIC3VdWc5udxVTUYLu/tGvaBrvcP8NCZ36G1rLW2YcbNkFdJkqQprZ9C5xLgpUk2T7I18MLunUm2AmZU1Xl0boHPadp3rarLmweObqUTPr8JvDXJpk2fJyTZcoz1vKJZk7krsAvww/W4pouBlzfj7AAcsh5jSJIkTXh9s6azqq5MciawHLgRuGhIl62BrybZjM4M4t807R9MMrtp+w5wNbCCzq3yK5ME+BWd2dGx+CFwIZ1b5W+pqt91hhqTLwHPBq4BfkTnVv/qsQ4iSZI00aVqpLvG2hBJtqqqu5rb/kuBA5r1ncOaPnN2zTz61I1X4BCrFhzes3NLkqSJJcmyqpo7mr59M9M5iZ3bPFX/cOCf1hU4JUmSJitDZ8uq6pBe1yBJktRr/fQgkSRJkiYpZzr7zJ47zWDAdZWSJGmScaZTkiRJrTN0SpIkqXWGTkmSJLXO0ClJkqTW+SBRn1l502pmzV/Us/P75fCSJKkNznRKkiSpdYZOSZIktc7QKUmSpNYZOiVJktQ6Q6ckSZJaZ+gcpSTHJ9mi6/15SR7Ry5okSZImCkPn6B0P/CF0VtXzq+r2HtYjSZI0Yfg9ncNIsiVwFvCnwDT4/+3dfZhedX3n8feHuAQoMVZ5MIGVKGIpEkhh8AFFHky7KAqrpUbLrkCtVG3V1aIbi2ux10UN1d1Soa2bsoJFLBSoSsUHBAEFijDhwRAeW4jSgCiKEUURwnf/uE/s7XQGJumcOXPP/X5d11w59++c3znf330lmc/8zu/cw3nAQuCyJA9U1cFJ1gIjwLbAF4Argf2BdcARVfWTJPsB/w/4cbP/FVW153SPR5IkqWvOdI7vUODeqtq7CYmnAPcCB1fVweMcvxvwl1X1fOAHwG827WcAb6mqFwMbJrpYkuOSjCYZ3fDw+ikdiCRJ0kxg6BzfamBpkpOTHFBVT5YE766qG5vtVcCiZr3nvKq6umn/1ESdq2plVY1U1cicbeb/x6uXJEmaYby9Po6quiPJvsArgQ8lufhJujzSt70B2BpIW/VJkiQNGmc6x5FkIfBwVX0S+AiwD/AQMG+y56iqB4GHkryoaXr9lBcqSZI0IJzpHN9i4MNJHgceBd4KvBj4QpL7JljXOZ43AX+T5MfA5YALNiVJ0lAydI6jqr4EfGlM8yhwat8xi5rNB4A9+9o/0tdnTVXtBZBkeXMOSZKkoWPobNdhSd5H733+JnBMt+VIkiR1w9DZoqo6Fzi36zokSZK65oNEkiRJap0znTPM4p3mM7risK7LkCRJmlLOdEqSJKl1hk5JkiS1ztApSZKk1hk6JUmS1DofJJphVq9bz6LlF3V2/bU+xCRJklrgTKckSZJaZ+iUJElS6wydkiRJap2hU5IkSa0byNCZ5GlJ3tbyNY5Jclqb15AkSRoWAxk6gacB/y50JpnTQS3jmkm1SJIkdW1QQ+cKYNckNya5LsllST4FrAZI8t+SXNvs/78bA2CSv04ymmRNkg9uPFmS/ZJcneSmpt+8ZtfCJF9McmeSP+s7/jeS/FOS65Ocl2Tbpn1tkg8kuRJYnuT6vj67JVnV/lsjSZI08wxq6FwO/EtVLQHeA7wAOKGq9kjyq8Ay4CXN/g3AUU2/E6pqBNgLODDJXkm2BM4F3llVewNLgZ80xy9pzrUYWJbkPyfZDng/sLSq9gFGgXf31fbTqnppVZ0ErE+ypGk/Fjhz6t8KSZKkmW+2fDj8tVV1d7P9cmBf4LokAFsD32n2vS7JcfTGvQDYAyjgvqq6DqCqfgjQ9L20qtY3r28BdqF3a38P4KrmmC2Bf+qr5dy+7dOBY5O8m154fcF4xTc1HQcw56nbb9YbIEmSNJPNltD5477tAJ+oqvf1H5Dk2cDxwH5V9WCSM4GtmuNrgvM+0re9gd77FeDLVfWGSdRyAfDHwFeAVVX1vfE6VNVKYCXA3AW7TVSLJEnSwBrU2+sPAfMm2HcpcGSSHQCSPD3JLsBT6QXC9Ul2BF7RHH8bvbWb+zXHz0vyRGH8GuAlSZ7bHL9NkueNd2BV/RT4EvDXwBmbMkBJkqTZZCBnOqvqe0muSnIzvfWX9/ftuyXJ+4GLk2wBPAr8flVdk+QGYA1wF3BVc/zPkiwDTk2ydXO+pU9w7e8mOQb4uyRzm+b3A3dM0OVs4LXAxZs/YkmSpMGWKu/mtinJ8cD8qvpfkzl+7oLdasHRp7Rc1cTWrjiss2tLkqTBkmRV85D2kxrImc5BkeTTwK7AIV3XIkmS1CVDZ4uq6jVd1yBJkjQTDOqDRJIkSRoghk5JkiS1ztvrM8zineYz6sM8kiRplnGmU5IkSa0zdEqSJKl1hk5JkiS1zjWdM8zqdetZtPyizq7vh8NLkqQ2ONMpSZKk1hk6JUmS1DpDpyRJklpn6JQkSVLrDJ2SJElq3awJnUlOTHJ8B9ddmOT8ZntJkldOdw2SJEkz3awJnVMhySZ/hFRV3VtVRzYvlwCGTkmSpDEGOnQmOSHJ7UkuAX6lads1yReTrErytSS7N+1nJvlY03ZHklc17cckOS/JPwIXJ9k2yaVJrk+yOskRzXEnJ3lb37VPTPKHSRYluTnJlsCfAMuS3JhkWZI7k2zfHL9Fkn9Ost30vkuSJEndG9gPh0+yL/B64NfojeN6YBWwEnhLVd2Z5IXAXwGHNN0WAQcCuwKXJXlu0/5iYK+q+n4z2/maqvphExCvSXIhcA5wSnM+gNcBh9IE96r6WZIPACNV9QdNjbsDRzX9lgI3VdUD44zlOOA4gDlP3X4q3h5JkqQZZWBDJ3AA8OmqehigCYZbAfsD5yXZeNzcvj5/X1WPA3cmuQvYvWn/clV9v9kO8KdJXgY8DuwE7FhVNyTZIclCYHvgwar6VpJFT1Djx4HP0gudvwOcMd5BVbWSXlhm7oLdapLjlyRJGhiDHDoBxga0LYAfVNWSSR6/8fWP+9qOohcq962qR5OspRdmAc4HjgSeSW/m84mLq7onyf1JDgFe2JxbkiRp6Azyms6vAq9JsnWSecCrgYeBu5P8FkB69u7r81vN2spdgecAt49z3vnAd5rAeTCwS9++c+jd0j+SXgAd6yFg3pi204FP0ptl3bDJo5QkSZoFBjZ0VtX1wLnAjcAFwNeaXUcBb0pyE7AGOKKv2+3AFcAX6K37/Ok4pz4bGEky2pzrtr5rrqEXKtdV1X3j9L0M2GPjg0RN24XAtkxwa12SJGkYDPTt9ao6CThpnF2HTtDlqqp615hznAmc2ff6AXoPFk10zcVjXq8F9my2vw/sN6bL3vQeILoNSZKkITXQoXOmS7IceCuu5ZQkSUNuaEJnVR3TwTVXACum+7qSJEkzzcCu6ZQkSdLgGJqZzkGxeKf5jK44rOsyJEmSppQznZIkSWqdoVOSJEmtM3RKkiSpdYZOSZIktc4HiWaY1evWs2j5RZ1df60PMUmSpBY40ylJkqTWGTolSZLUOkOnJEmSWmfolCRJUusMnZIkSWqdoVOSJEmtM3RKkiSpdYbOzZDkjUm+keSmJGcl2SXJpU3bpUme1Rx3ZpKPJrk6yV1Jjuy6dkmSpC4YOjdRkucDJwCHVNXewDuB04C/raq9gLOBj/Z1WQC8FHgVsGKCcx6XZDTJ6IaH17davyRJUhcMnZvuEOD8qnoAoKq+D7wY+FSz/yx6IXOjz1TV41V1C7DjeCesqpVVNVJVI3O2md9i6ZIkSd0wdG66APUkx/Tvf2RMX0mSpKFj6Nx0lwKvS/IMgCRPB64GXt/sPwq4sqPaJEmSZqSndF3AoKmqNUlOAq5IsgG4AXgH8PEk7wG+CxzbZY2SJEkzjaFzM1TVJ4BPjGk+ZJzjjhnzetsWy5IkSZqxvL0uSZKk1hk6JUmS1DpDpyRJklrnms4ZZvFO8xldcVjXZUiSJE0pZzolSZLUOkOnJEmSWmfolCRJUusMnZIkSWqdDxLNMKvXrWfR8os6u/5aH2KSJEktcKZTkiRJrTN0SpIkqXWGTkmSJLXO0ClJkqTWDVXoTHJ18+eiJDc32wcl+dxmnu/0JHtMZY2SJEmz0VA9vV5V+0/x+X53Ks8nSZI0Ww3bTOePnmT/iUmO73t9czMr+ktJLkpyU9O2rNl/eZKRZvsNSVY3+0/uv2aSk5q+1yTZsa3xSZIkzVRDFTr/Aw4F7q2qvatqT+CL/TuTLAROBg4BlgD7Jfmvze5fAq6pqr2BrwJvnr6yJUmSZgZD5+SsBpYmOTnJAVW1fsz+/YDLq+q7VfUYcDbwsmbfz4CNa0ZXAYvGnjzJcUlGk4xueHjsqSVJkgafofMXPcYvvidbAVTVHcC+9MLnh5J8YEy/PME5H62qarY3MM462qpaWVUjVTUyZ5v5m128JEnSTGXo/EVrgX0AkuwDPLvZXgg8XFWfBD6y8Zg+XwcOTLJdkjnAG4ArpqtoSZKkmW6onl6fhAuANya5EbgOuKNpXwx8OMnjwKPAW/s7VdV9Sd4HXEZv1vPzVfXZ6StbkiRpZsu/3fnVTDB3wW614OhTOrv+2hWHdXZtSZI0WJKsqqqRyRzr7XVJkiS1ztApSZKk1hk6JUmS1DpDpyRJklrn0+szzOKd5jPqwzySJGmWcaZTkiRJrTN0SpIkqXWGTkmSJLXONZ0zzOp161m0/KLOru+Hw0uSpDY40ylJkqTWGTolSZLUOkOnJEmSWmfolCRJUusMnZIkSWrdUIfOJAuTnN9sH5Tkc832MUlOe5K+hydZPh11SpIkDbqh/sikqroXOHIz+14IXDi1FUmSJM1OQzPTmUCbLYAAAAkgSURBVOTkJG/re31ikj9McvOT9Ht1kq8nuSHJJUl2bNp/Phua5MwkH01ydZK7khzZtJ+V5Ii+c52d5PB2RihJkjRzDU3oBM4BlvW9fh1w3ST6XQm8qKp+rTnHeyc4bgHwUuBVwIqm7XTgWIAk84H9gc+P7ZjkuCSjSUY3PLx+EiVJkiQNlqG5vV5VNyTZIclCYHvgQeBbk+i6M3BukgXAlsDdExz3map6HLhl42xoVV2R5C+T7AC8Frigqh4bp7aVwEqAuQt2q00dmyRJ0kw3TDOdAOfTW8O5jN6s5WScCpxWVYuB3wO2muC4R/q207d9FnAUvRnPMzapWkmSpFliaGY6G+cAfwNsBxwIzJ1En/nAumb76M245pnAtcC3q2rNZvSXJEkaeEM109mEvnnAuqq6b5LdTgTOS/I14IHNuOb9wK04yylJkoZYqlxC2KYk2wCrgX2q6kmfEpq7YLdacPQp7Rc2gbUrDuvs2pIkabAkWVVVI5M5dqhmOqdbkqXAbcCpkwmckiRJs9WwremcVlV1CfCsruuQJEnqmjOdkiRJap0znTPM4p3mM+q6SkmSNMs40ylJkqTWGTolSZLUOkOnJEmSWmfolCRJUusMnZIkSWqdoVOSJEmtM3RKkiSpdYZOSZIktc7QKUmSpNYZOiVJktQ6Q6ckSZJaZ+iUJElS6wydkiRJap2hU5IkSa0zdEqSJKl1hk5JkiS1ztApSZKk1hk6JUmS1DpDpyRJklpn6JQkSVLrDJ2SJElqnaFTkiRJrTN0SpIkqXWGTkmSJLUuVdV1DeqT5CHg9q7r6Mh2wANdF9GBYR03DO/Yh3XcMLxjH9Zxw/COfVjGvUtVbT+ZA5/SdiXaZLdX1UjXRXQhyegwjn1Yxw3DO/ZhHTcM79iHddwwvGMf1nE/EW+vS5IkqXWGTkmSJLXO0DnzrOy6gA4N69iHddwwvGMf1nHD8I59WMcNwzv2YR33hHyQSJIkSa1zplOSJEmtM3RKkiSpdYZOSZIktc7P6exYkt2BI4CdgALuBS6sqls7LWwaNGPfCfh6Vf2or/3Qqvpid5VNryR/W1Vv7LqOtiV5IXBrVf0wydbAcmAf4BbgT6tqfacFtiTJlsDrgXur6pIkvw3sD9wKrKyqRzstUJKmiQ8SdSjJ/wTeAJwD/GvTvDO9b1DnVNWKrmprW5J3AL9P7xvvEuCdVfXZZt/1VbVPl/W1JcmFY5uAg4GvAFTV4dNe1DRJsgbYu6oeS7ISeBg4H3h50/7aTgtsSZKz6f2Avw3wA2Bb4B/ojTtVdXSH5UnStHGms1tvAp4/dqYjyf8B1gCzNnQCbwb2raofJVkEnJ9kUVX9Bb0gNlvtTG9m73R6M9sBRoD/3WVR02SLqnqs2R7p+8HiyiQ3dlXUNFhcVXsleQqwDlhYVRuSfBK4qePapNYl2aGqvtN1HV1I8oyq+l7XdcwUruns1uPAwnHaFzT7ZrM5G2+pV9Va4CDgFU3gns2hcwRYBZwArK+qy4GfVNUVVXVFp5W17+YkxzbbNyUZAUjyPGA232LeornFPo/ebOf8pn0u8J86q2oaJJmfZEWS25J8r/m6tWl7Wtf1tSnJU5N8KMlZzZKK/n1/1VVdbUvy9DFfzwCuTfLLSZ7edX1tav5eb9dsjyS5C/h6km8mObDj8mYEb693KMmhwGnAncA9TfOzgOcCfzCb1zUm+Qrw7qq6sa/tKcDHgaOqak5nxU2DJDsDfw7cDxxeVc/quKTWJZkP/AVwAPAAvfWc9zRf76iqWTnrl+RdwNuBOfRmtI8A7gJeBJxfVR/ssLxWJfkSvaUjn6iqbzdtzwSOBpZW1a93WV+bklxA7//2a4DfofeD1W9X1SOzfAnR48A3xzTvTG8JWVXVc6a/qumRZHVVLW62LwPeW1XXNT9Yf8rfw27o7FySLYAX0HugJvT+YV5XVRs6LaxlTeh6bOM3ojH7XlJVV3VQ1rRLchjwkqr6o65rmS5J5gHPobe851+r6v6OS2pdkoUAVXVvM8O3FPhWVV3bbWXtSnJ7Vf3Kpu6bDZLcWFVL+l6fALwSOBz48iwOncfT+/v9nqpa3bTdXVXP7ray9iW5DdizWbd+TVW9qG/fzwPpMDN0SpJakeRi4BJ6M533N207AscAv15VSzssr1VJbqW3Zv/xvrajgfcC21bVLp0V17K+Ozn3AH8M3DSbZzg3SvJ24NX0nsd4GfA0/u2hwedU1X/vsLwZwdApSWpFkl+m99FYRwA7NM33AxcCK6rqwa5qa1uSPwMurqpLxrQfCpxaVbt1U9n0SfJqeuvXF1XVM7uuZzokOQh4K/A8endz7gE+A3y870HKoWXolCRNuyTHVtUZXdfRhWEae/OZvLtW1c3DNO6xhnns/QydkqRpl+Rbw/AA3XiGdezDOm4Y7rH383M6JUmtSPKNiXYBO05nLdNtWMc+rOOG4R77ZBk6JUlt2RH4L8DYtZsBrp7+cqbVsI59WMcNwz32STF0SpLa8jl6T2r/u984leTy6S9nWg3r2Id13DDcY58U13RKkiSpdf4aTEmSJLXO0ClJkqTWGTolacAkeWaSc5L8S5Jbkny++f3OU3X+g5LsP1XnkyQwdErSQEkS4NPA5VW1a1XtAfwRU/uRLAcBhk5JU8rQKUmD5WDg0ar62MaG5mnZK5N8OMnNSVYnWQY/n7X83MZjk5yW5Jhme22SDya5vumze5JFwFuAdyW5MckB0zg2SbOYH5kkSYNlT2DVOO2vBZYAewPbAdcl+eokzvdAVe2T5G3A8VX1u0k+Bvyoqj4yZVVLGnrOdErS7PBS4O+qakNV3Q9cAew3iX7/0Py5CljUUm2SZOiUpAGzBth3nPZMcPxj/OL/9VuN2f9I8+cGvPslqUWGTkkaLF8B5iZ588aGJPvR+9V7y5LMSbI98DLgWuCbwB5J5iaZD7x8Etd4CJg39aVLGmb+VCtJA6SqKslrgFOSLAd+CqwF/gewLXATUMB7q+rbAEn+HvgGcCdwwyQu84/A+UmOAN5eVV+b8oFIGjr+GkxJkiS1ztvrkiRJap2hU5IkSa0zdEqSJKl1hk5JkiS1ztApSZKk1hk6JUmS1DpDpyRJklr3/wHPXzea28Z7CgAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#######\n",
    "# a well-labeled barplot\n",
    "plt.figure(figsize = (10, 10)) # make it medium and square\n",
    "plt.barh(finalcount.index, finalcount) # tell the barplot what to put on the x and y axes. This time we're using 'barh' for a horizontal bar plot.\n",
    "plt.xticks(rotation=90) # this tells Python to 'rotate' the labels on the x axis so that the words fit on the screen\n",
    "plt.ylabel('Word') # label\n",
    "plt.xlabel('Count') # label\n",
    "plt.suptitle('Words for Wrongdoing in Jane Austen') #This line labels the title\n",
    "plt.gca().invert_yaxis() # reverse the order of the y axis\n",
    "plt.savefig('austen-wrongdoing.png', dpi = 300) # save the figure in your folder\n",
    "plt.show() # display the figure below"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Load Congress Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "congress = pd.read_csv(\"congress-just-1967-2010.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "congress1 = congress[congress['year'] == 1967]\n",
    "congress2 = congress[congress['year'] == 2010]"
   ]
  },
  {
   "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>Unnamed: 0</th>\n",
       "      <th>Unnamed: 0.1</th>\n",
       "      <th>speech</th>\n",
       "      <th>date</th>\n",
       "      <th>speaker</th>\n",
       "      <th>word_count</th>\n",
       "      <th>year</th>\n",
       "      <th>month</th>\n",
       "      <th>month_year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>185175</th>\n",
       "      <td>5926838</td>\n",
       "      <td>5926838</td>\n",
       "      <td>The clerk will please read a communication to ...</td>\n",
       "      <td>2010-01-05</td>\n",
       "      <td>The PRESIDING OFFICER</td>\n",
       "      <td>27</td>\n",
       "      <td>2010</td>\n",
       "      <td>1</td>\n",
       "      <td>2010-01-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>185176</th>\n",
       "      <td>5926839</td>\n",
       "      <td>5926839</td>\n",
       "      <td>Under the previous order. the Senate stands ad...</td>\n",
       "      <td>2010-01-05</td>\n",
       "      <td>The ACTING PRESIDENT pro tempore</td>\n",
       "      <td>83</td>\n",
       "      <td>2010</td>\n",
       "      <td>1</td>\n",
       "      <td>2010-01-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>185177</th>\n",
       "      <td>5926840</td>\n",
       "      <td>5926840</td>\n",
       "      <td>The Chair will lead the House in the Pledge of...</td>\n",
       "      <td>2010-01-05</td>\n",
       "      <td>The SPEAKER pro tempore</td>\n",
       "      <td>11</td>\n",
       "      <td>2010</td>\n",
       "      <td>1</td>\n",
       "      <td>2010-01-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>185178</th>\n",
       "      <td>5926841</td>\n",
       "      <td>5926841</td>\n",
       "      <td>Pursuant to clause 4 of rule I. the following ...</td>\n",
       "      <td>2010-01-05</td>\n",
       "      <td>The SPEAKER pro tempore</td>\n",
       "      <td>40</td>\n",
       "      <td>2010</td>\n",
       "      <td>1</td>\n",
       "      <td>2010-01-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>185179</th>\n",
       "      <td>5926842</td>\n",
       "      <td>5926842</td>\n",
       "      <td>Pursuant to clause 5(d) of rule XX. the Chair ...</td>\n",
       "      <td>2010-01-05</td>\n",
       "      <td>The SPEAKER pro tempore</td>\n",
       "      <td>21</td>\n",
       "      <td>2010</td>\n",
       "      <td>1</td>\n",
       "      <td>2010-01-01</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        Unnamed: 0  Unnamed: 0.1  \\\n",
       "185175     5926838       5926838   \n",
       "185176     5926839       5926839   \n",
       "185177     5926840       5926840   \n",
       "185178     5926841       5926841   \n",
       "185179     5926842       5926842   \n",
       "\n",
       "                                                   speech        date  \\\n",
       "185175  The clerk will please read a communication to ...  2010-01-05   \n",
       "185176  Under the previous order. the Senate stands ad...  2010-01-05   \n",
       "185177  The Chair will lead the House in the Pledge of...  2010-01-05   \n",
       "185178  Pursuant to clause 4 of rule I. the following ...  2010-01-05   \n",
       "185179  Pursuant to clause 5(d) of rule XX. the Chair ...  2010-01-05   \n",
       "\n",
       "                                 speaker  word_count  year  month  month_year  \n",
       "185175             The PRESIDING OFFICER          27  2010      1  2010-01-01  \n",
       "185176  The ACTING PRESIDENT pro tempore          83  2010      1  2010-01-01  \n",
       "185177           The SPEAKER pro tempore          11  2010      1  2010-01-01  \n",
       "185178           The SPEAKER pro tempore          40  2010      1  2010-01-01  \n",
       "185179           The SPEAKER pro tempore          21  2010      1  2010-01-01  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "congress2.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Clean the Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'those who do not enjoy the privilege of the floor will please retire from the chamber mr president on the basis of an agreement reached on both sides '"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# clean up speeches from 1967\n",
    "congressspeeches1 = []\n",
    "for speech in congress1['speech']:\n",
    "    speech = re.sub('[\\\",.;:?([)\\]_*]', '', speech) # strip punctuation\n",
    "    speech = re.sub('-', ' ', speech) # replace hyphens with spaces\n",
    "    speech = speech.lower()\n",
    "    congressspeeches1.append(speech)\n",
    "congressspeeches1 = ' '.join(congressspeeches1)\n",
    "congressspeeches1[:150]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'the clerk will please read a communication to the senate from the president pro mr warner thereupon assumed the tempore  chair as acting president pro'"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# clean up speeches from 2010\n",
    "congressspeeches2 = []\n",
    "for speech in congress2['speech']:\n",
    "    speech = re.sub('[\\\",.;:?([)\\]_*]', '', speech) # strip punctuation\n",
    "    speech = re.sub('-', ' ', speech) # replace hyphens with spaces\n",
    "    speech = speech.lower() # lower case 'em '\n",
    "    congressspeeches2.append(speech) # save the speeches for later\n",
    "congressspeeches2 = ' '.join(congressspeeches2)\n",
    "congressspeeches2[:150]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Try Using the Austen Code to Find Vocab Matches in Congress\n",
    "\n",
    "The code above -- the code we have been using all week to find matches -- could *theoretically* run on the Congress dataset.  But it might crash the machine. Let's see what happens.  If the lines below crash, try following along with the lines below, which do the same thing another way."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# search for matched words\n",
    "congresswords1 = congressspeeches1.split()\n",
    "matchedwords = []\n",
    "for w in congresswords1:\n",
    "    w = wn.morphy(w) # lemmatize it\n",
    "    if w in vocab:\n",
    "        matchedwords.append(w)\n",
    "        \n",
    "# find the bigrams \n",
    "matchedbigrams = []\n",
    "bigrams = TextBlob(congressspeeches1).ngrams(n=2)\n",
    "for bigram in bigrams: # move through each line of the *bigrams* list\n",
    "        if wn.morphy(bigram[0]) is not None and wn.morphy(bigram[1]) is not None:\n",
    "            bigram2 = wn.morphy(bigram[0]) + ' ' + wn.morphy(bigram[1]) # call the first word, a space, and the second word into a new string\n",
    "            if bigram2 in cleanwordnetbigrams:\n",
    "                matchedbigrams.append(bigram2) # save the string \n",
    "                \n",
    "# put it all together\n",
    "allvocabcongress1 = matchedwords+matchedbigrams\n",
    "\n",
    "# stopword out ambiguous words\n",
    "finallist = []\n",
    "for v in allcontrolledvocab:\n",
    "    if v not in ambiguouswords:\n",
    "        finallist.append(v)\n",
    "finalcount = pd.Series.value_counts(finallist)\n",
    "finalcount[:5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Your code likely crashed.  \n",
    "\n",
    "If you received a message about the 'kernel restarting,' that means that you have lost all the variables you created up to this point.\n",
    "\n",
    "***We're going to keep going.***\n",
    "\n",
    "1) Next, run the code UP TO this block of code.  \n",
    "2) Skip the block of code immediately above these instructions -- the one that crashed your kernel.\n",
    "3) Continue on running the code below."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Find Vocab Matches in Congress More Efficiently"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Congress is bigger than Austen!  In order to get our code to run quickly, we have to use slightly more efficient code.  \n",
    "\n",
    "In the weeks that follow, we will be learning a number of shortcuts to do things we've already learned in a more \"efficient\" manner.  This means that the code will talk to Python about how to allocate computer power in such a way that the code will run more swiftly and effectively over the data in question.\n",
    "\n",
    "Let's introduce a concept that will help us understand computational efficiency:\n",
    "   * Writing Data Takes Time\n",
    "   * More efficient code stores the results as it goes\n",
    "   \n",
    "The code for matching we've written to date requires the computer to call lists of words again and again: calling up the entire list of words in the original text to execute a \".append().\"  When we're working with a million words, the .append() command slows us down. The overhead of creating and extending the list slows the process down.  \n",
    "\n",
    "Computers work more efficiently when they work only on the precise data required.  \n",
    "\n",
    "\n",
    "#### Working in parallel with list comprehension.\n",
    "\n",
    "Our first lesson in efficient code is 'list comprehension.'  List comprehension helps us to write a for-loop in such a way that the computer doesn't have to use **.append()** every time. \n",
    "\n",
    "Here is a for loop that we've used before:\n",
    "\n",
    "    for word in words:\n",
    "        if word in vocab:\n",
    "            do something.\n",
    "\n",
    "Now, let's write the for loop so that it can be run in parallel.\n",
    "\n",
    "The lines below do the same as those above, but we use 'list comprehension' grammar:\n",
    "    \n",
    "    [word for word in words if word in vocab]\n",
    "\n",
    "The brackets -- **[ ]** -- tell Python that we are making a list.  The grammar \"for blank in blank\" tells Python that we want a for-loop type of program.  Instead of using **.append()** every time to store the results in a new list, we simply harvest the results and make a new list as we go.  \n",
    "\n",
    "With some practice, you can become adept at translating traditional for-loops into list comprehension loops, which are more efficient.\n",
    "\n",
    "Don't fret too much about this; we'll be learning several new tricks for efficient code in the coming weeks.  So long as you can recognize list comprehension and look up how to do it, you're good to go."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Use list comprehension to find matched words\n",
    "\n",
    "First, fine the controlled vocab match in words and bigrams for 1967 and plot the top instances."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*This may be slow with smaller memory, but it will be fast on JupyterLab Sessions run with Memory > 60GB*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['gore', 'vice', 'vice', 'vice', 'fall']"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# search for matched words\n",
    "setvocab = set(vocab)\n",
    "congresswords1 = congressspeeches1.split()\n",
    "matchedwords1 = [wn.morphy(word) for word in congresswords1 if wn.morphy(word) in setvocab]\n",
    "matchedwords1[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# we'll use a slightly different method for finding ngrams because it's faster with big data than TextBlob.ngrams()\n",
    "# these lines define a new function -- zipngram -- which creates ngrams of n length. \n",
    "def zipngram(text, n=2):\n",
    "    words = text.split()\n",
    "    return zip(*[words[i:] for i in range(n)])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['aggravated assault',\n",
       " 'criminal offense',\n",
       " 'juvenile delinquency',\n",
       " 'armed robbery',\n",
       " 'juvenile delinquency']"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# search for matched bigrams\n",
    "bigrams = zipngram(congressspeeches1)\n",
    "matchedbigrams1= [' '.join(bigram) for bigram in bigrams if ' '.join(bigram) in setvocab]\n",
    "matchedbigrams1[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "# put it all together\n",
    "allcontrolledvocab1 = matchedwords1 + matchedbigrams1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# a big list of ambiguous words\n",
    "ambiguouswords = ['infringement', 'setup', 'mischief', 'foul', 'illusion','encroachment', 'misdemeanor','unfairness', 'pretense','sophistication', 'deceit',\n",
    "                  'disservice','fabrication', 'equivocation','falsehood','usurpation','violation', 'injury', 'injustice', 'carnage', 'betrayal','evil', \n",
    "                  'hell', 'offense','deviation', 'impropriety','indirection','pretending','duplicity', 'delusion','trickery','shrinkage','transgression',\n",
    "                  'spoiling', 'indecency','depravity','sin','slay','butchery','molestation','mortal sin','guile','abomination','malicious mischief',\n",
    "                  'treason','treachery','inroad','offence','foul play','perpetration','wile','iniquity','committal','brutalization','infliction','falsify','hooliganism','impersonation',\n",
    "                    'sinning','fakery', 'hustle', 'mangle','intemperance', 'foul ball','fratricide','monkey business','terrorization','pilferage','gyp',\n",
    "                  'pyramiding','obscenity','holdup', 'wickedness','nonfeasance','abnormality','law breaking','mutilate','caper','ill turn','frame up',\n",
    "                  'prevarication','thuggery','immorality','perversion','rascality','quackery','double dealing','con','fraud','shenanigan','roguery','high treason',\n",
    "                  'cheating','off','setup,' 'foul', 'wrong', 'liberty', 'attempt', 'liquidation', 'petty', 'battery', 'gaming', 'intrusion','raid','misfeasance',\n",
    "                  'attack','assault','sting','indiscretion','vice','crime','snatch','infraction','misapplication','kill','irregularity','flimflam','chicanery',\n",
    "                  'sellout','twisting','blaze','straining','misbehavior','warp','spoilage','tort','simulation','turpitude','dereliction','mayhem', 'dishonesty',\n",
    "                  'cheat', 'trick','deception','skimming', 'distort','bluff', 'distortion', 'misrepresentation','execute','gore', 'job', 'technical',\n",
    "                  'damage', 'elimination', 'take in', 'burke', 'remove', 'fall', 'familiarity', 'pretence', 'maintenance', 'spoil', 'commission', \n",
    "                  'protection', 'execution', 'tribute', 'hit', 'dispatch', 'play', 'rolling']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "# stopword out ambiguous words \n",
    "finallist1 = [w for w in allcontrolledvocab1 if w not in ambiguouswords]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "murder                  670\n",
       "delinquency             668\n",
       "slaughter               391\n",
       "juvenile delinquency    381\n",
       "robbery                 273\n",
       "dtype: int64"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# count totals\n",
    "finalcount1 = pd.Series.value_counts(finallist1)[:50]\n",
    "finalcount1[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 360x1440 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# a well-labeled barplot\n",
    "plt.figure(figsize = (5, 20)) # make it tall and thin\n",
    "plt.tight_layout() # adjust the margins so that the title is near the plot \n",
    "plt.subplots_adjust(top=0.95)\n",
    "plt.gcf().subplots_adjust(left=0.35) # pad the left margin so that the labels don't get cut off\n",
    "plt.barh(finalcount1.index, finalcount1) # tell the barplot what to put on the x and y axes. This time we're using 'barh' for a horizontal bar plot.\n",
    "plt.xticks(rotation=90) # this tells Python to 'rotate' the labels on the x axis so that the words fit on the screen\n",
    "plt.ylabel('Word') # label\n",
    "plt.xlabel('Count') # label\n",
    "plt.suptitle('Words for Wrongdoing in Congress in 1967', fontsize=20) #This line labels the title\n",
    "plt.gca().invert_yaxis() # reverse the order of the y axis\n",
    "plt.savefig('congress-wrongdoing-1967.png', dpi = 300) # save the figure in your folder\n",
    "plt.show() # display the figure below"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Do the same thing for 2010."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# search for matched words\n",
    "setvocab = set(vocab)\n",
    "congresswords2 = congressspeeches2.split()\n",
    "matchedwords2 = [wn.morphy(word) for word in congresswords2 if wn.morphy(word) in setvocab]\n",
    "matchedwords2[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# search for matched bigrams\n",
    "bigrams = zipngram(congressspeeches2)\n",
    "matchedbigrams2= [' '.join(bigram) for bigram in bigrams if ' '.join(bigram) in setvocab]\n",
    "matchedbigrams2[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# put it all together\n",
    "allcontrolledvocab2 = matchedwords2 + matchedbigrams2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# stopword out ambiguous words \n",
    "finallist2 = [w for w in allcontrolledvocab2 if w not in ambiguouswords]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# count totals\n",
    "finalcount2 = pd.Series.value_counts(finallist2)[:50]\n",
    "finalcount2[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# a well-labeled barplot\n",
    "plt.figure(figsize = (5, 20)) # make it tall and thin\n",
    "plt.barh(finalcount2.index, finalcount2) # tell the barplot what to put on the x and y axes. This time we're using 'barh' for a horizontal bar plot.\n",
    "plt.tight_layout() # adjust the margins so that the title is near the plot \n",
    "plt.subplots_adjust(top=0.95)\n",
    "plt.gcf().subplots_adjust(left=0.35) # pad the left margin so that the labels don't get cut off\n",
    "plt.xticks(rotation=90) # this tells Python to 'rotate' the labels on the x axis so that the words fit on the screen\n",
    "plt.ylabel('Word') # label\n",
    "plt.xlabel('Count') # label\n",
    "plt.suptitle('Words for Wrongdoing in Congress in 2010', fontsize=20) #This line labels the title\n",
    "plt.gca().invert_yaxis() # reverse the order of the y axis\n",
    "plt.savefig('congress-wrongdoing-2010.png', dpi = 300) # save the figure in your folder\n",
    "plt.show() # display the figure below"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Assignment\n",
    "\n",
    "Use the visualizations included here to answer the following essay questions.  Turn in a word document with well-formatted answers.\n",
    "\n",
    "1) What kinds of misdeeds concern Austen? Are these actually crimes? Be specific. Write about as many words as possible. Embed the visualization in a Word Document. Entitle this section, \"Misdeeds in Jane Austen.\"\n",
    "\n",
    "2) Compare and contrast wrongdoing in Jane Austen and wrongdoing in Congress.  Why were a novelist from 1800 and lawmakers after 1967 concerned with different offenses? Entitle this section, \"Congress is not Jane Austen.\"\n",
    "\n",
    "3) What kinds of crimes concerned Congress in 1967?  Be specific. Write about as many words as possible. Embed the visualization in a Word Document. Entitle this section, \"Congress in 1967.\"\n",
    "\n",
    "4) What kinds of crimes concerned Congress in 2010?  Be specific. Write about as many words as possible. Embed the visualization in a Word Document. Entitle this section, \"Congress in 2010.\"\n",
    "\n",
    "5) Under the heading \"Explaining the Contrast,\" write a paragraph comparing and contrasting wrongdoing in Congress in 1967 with wrongdoing in Congress in 2010. Be specific. Write about as many words as possible. What might explain this historical change of focus?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.9.10 (main, Feb 25 2022, 18:07:24) \n[Clang 13.0.0 (clang-1300.0.29.30)]"
  },
  "vscode": {
   "interpreter": {
    "hash": "aee8b7b246df8f9039afb4144a1f6fd8d2ca17a180786b69acc140d282b71a49"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}