LLR_categroies.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import time\n",
    "import collections\n",
    "import pickle\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Log Likelihood Ratio for words following \"he\" or \"she\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load category-specific bigram counters\n",
    "\n",
    "with open('data/bigrams/female_sports.pkl', 'rb') as infile:\n",
    "    female_sports = pickle.load(infile)\n",
    "# with open('data/bigrams/female_monarch.pkl', 'rb') as infile:\n",
    "#     female_monarch = pickle.load(infile)\n",
    "with open('data/bigrams/female_scientist.pkl', 'rb') as infile:\n",
    "    female_scientist = pickle.load(infile)\n",
    "with open('data/bigrams/female_artists.pkl', 'rb') as infile:\n",
    "    female_artists = pickle.load(infile)\n",
    "with open('data/bigrams/female_programming.pkl', 'rb') as infile:\n",
    "    female_programming = pickle.load(infile)\n",
    "with open('data/bigrams/male_programming.pkl', 'rb') as infile:\n",
    "    male_programming = pickle.load(infile)\n",
    "with open('data/bigrams/male_scientist.pkl', 'rb') as infile:\n",
    "    male_scientist = pickle.load(infile)\n",
    "with open('data/bigrams/male_sports.pkl', 'rb') as infile:\n",
    "    male_sports = pickle.load(infile)\n",
    "with open('data/bigrams/male_artists.pkl', 'rb') as infile:\n",
    "    male_artists = pickle.load(infile)\n",
    "# with open('data/bigrams/male_monarch.pkl', 'rb') as infile:\n",
    "#     male_monarch = pickle.load(infile)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load category-specific unigram counters\n",
    "with open('data/bigrams/female_sports_u.pkl', 'rb') as infile:\n",
    "    female_sports_u = pickle.load(infile)\n",
    "# with open('data/bigrams/female_monarch_u.pkl', 'rb') as infile:\n",
    "#     female_monarch_u = pickle.load(infile)\n",
    "with open('data/bigrams/female_scientist_u.pkl', 'rb') as infile:\n",
    "    female_scientist_u = pickle.load(infile)\n",
    "with open('data/bigrams/female_artists_u.pkl', 'rb') as infile:\n",
    "    female_artists_u = pickle.load(infile)\n",
    "with open('data/bigrams/female_programming_u.pkl', 'rb') as infile:\n",
    "    female_programming_u = pickle.load(infile)\n",
    "with open('data/bigrams/male_programming_u.pkl', 'rb') as infile:\n",
    "    male_programming_u = pickle.load(infile)\n",
    "with open('data/bigrams/male_scientist_u.pkl', 'rb') as infile:\n",
    "    male_scientist_u = pickle.load(infile)\n",
    "with open('data/bigrams/male_sports_u.pkl', 'rb') as infile:\n",
    "    male_sports_u = pickle.load(infile)\n",
    "with open('data/bigrams/male_artists_u.pkl', 'rb') as infile:\n",
    "    male_artists_u = pickle.load(infile)\n",
    "# with open('data/bigrams/male_monarch_u.pkl', 'rb') as infile:\n",
    "#     male_monarch_u = pickle.load(infile)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# count of x, count of y, count of xy\n",
    "def pmi(cx, cy, cxy, n):\n",
    "    if cy == 0:\n",
    "        return 0\n",
    "    if cxy == 0:\n",
    "        return -100\n",
    "    else:\n",
    "        return np.log2(n) + np.log2(cxy) - (np.log2(cx) + np.log2(cy))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Category: Scientist"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2840901"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Number of words in corpus of women's SCIENTIST bios: \n",
    "n_women = sum(female_scientist_u.values())\n",
    "n_women  # 2840901"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "12296493"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Number of words in corpus of men's SCIENTIST bios: \n",
    "n_men = sum(male_scientist_u.values())\n",
    "n_men  # 12296493"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "# bigrams beginning with he or she, filtered by words appearing at least 75 times.\n",
    "\n",
    "she = collections.Counter({ k[1]: v for k, v in female_scientist.items() if v >= 75 and k[0] == \"she\" })\n",
    "he = collections.Counter({ k[1]: v for k, v in male_scientist.items() if v >= 75 and k[0] == \"he\" })"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "49"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(she)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "19504"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sum(she.values())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "149"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(he)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "90133"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sum(he.values())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of times 'he' appears in male bios: 106235\n",
      "Number of times 'she' appears in female bios: 26029\n"
     ]
    }
   ],
   "source": [
    "he_count = male_scientist_u['he']\n",
    "she_count = female_scientist_u['she']\n",
    "print(\"Number of times 'he' appears in male bios:\", he_count )\n",
    "print(\"Number of times 'she' appears in female bios:\", she_count)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Look up counts for all words in \"he\" counter not in \"she\" counter and vice versa\n",
    "for word in he:\n",
    "    if word not in she:\n",
    "        she[word] = female_scientist[(\"she\", word)]\n",
    "for word in she:\n",
    "    if word not in he:\n",
    "        he[word] = male_scientist[(\"he\", word)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "columns = ['Word following he/she', \"she+word count\"]\n",
    "df = pd.DataFrame([(k,v) for k,v in she.items()], columns=columns)\n",
    "df.set_index(['Word following he/she'], inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "columns2 = ['Word following he/she', \"he+word count\",]\n",
    "df2 = pd.DataFrame([(k,v) for k,v in he.items()], columns=columns2)\n",
    "df2.set_index(['Word following he/she'], inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.concat([df,df2], axis=1)\n",
    "df['total count'] = df.fillna(0)[\"she+word count\"] + df.fillna(0)[\"he+word count\"]\n",
    "df.sort_values(by=['total count'], ascending=False, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "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>she+word count</th>\n",
       "      <th>he+word count</th>\n",
       "      <th>total count</th>\n",
       "      <th>PMI she+word</th>\n",
       "      <th>PMI he+word</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>was</th>\n",
       "      <td>5683</td>\n",
       "      <td>22157</td>\n",
       "      <td>27840</td>\n",
       "      <td>5.027899</td>\n",
       "      <td>4.945018</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>is</th>\n",
       "      <td>1663</td>\n",
       "      <td>7190</td>\n",
       "      <td>8853</td>\n",
       "      <td>4.442270</td>\n",
       "      <td>4.281971</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>has</th>\n",
       "      <td>1188</td>\n",
       "      <td>4871</td>\n",
       "      <td>6059</td>\n",
       "      <td>5.091482</td>\n",
       "      <td>5.075041</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>also</th>\n",
       "      <td>830</td>\n",
       "      <td>3692</td>\n",
       "      <td>4522</td>\n",
       "      <td>4.579070</td>\n",
       "      <td>4.479146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>had</th>\n",
       "      <td>704</td>\n",
       "      <td>2970</td>\n",
       "      <td>3674</td>\n",
       "      <td>4.546522</td>\n",
       "      <td>4.517743</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>became</th>\n",
       "      <td>754</td>\n",
       "      <td>2906</td>\n",
       "      <td>3660</td>\n",
       "      <td>5.402951</td>\n",
       "      <td>5.338857</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>received</th>\n",
       "      <td>770</td>\n",
       "      <td>2833</td>\n",
       "      <td>3603</td>\n",
       "      <td>5.549835</td>\n",
       "      <td>5.567863</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>worked</th>\n",
       "      <td>511</td>\n",
       "      <td>1896</td>\n",
       "      <td>2407</td>\n",
       "      <td>5.428528</td>\n",
       "      <td>5.458474</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>served</th>\n",
       "      <td>294</td>\n",
       "      <td>1497</td>\n",
       "      <td>1791</td>\n",
       "      <td>5.157580</td>\n",
       "      <td>5.317230</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>studied</th>\n",
       "      <td>311</td>\n",
       "      <td>1193</td>\n",
       "      <td>1504</td>\n",
       "      <td>5.581214</td>\n",
       "      <td>5.529775</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          she+word count  he+word count  total count  PMI she+word  \\\n",
       "was                 5683          22157        27840      5.027899   \n",
       "is                  1663           7190         8853      4.442270   \n",
       "has                 1188           4871         6059      5.091482   \n",
       "also                 830           3692         4522      4.579070   \n",
       "had                  704           2970         3674      4.546522   \n",
       "became               754           2906         3660      5.402951   \n",
       "received             770           2833         3603      5.549835   \n",
       "worked               511           1896         2407      5.428528   \n",
       "served               294           1497         1791      5.157580   \n",
       "studied              311           1193         1504      5.581214   \n",
       "\n",
       "          PMI he+word  \n",
       "was          4.945018  \n",
       "is           4.281971  \n",
       "has          5.075041  \n",
       "also         4.479146  \n",
       "had          4.517743  \n",
       "became       5.338857  \n",
       "received     5.567863  \n",
       "worked       5.458474  \n",
       "served       5.317230  \n",
       "studied      5.529775  "
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['PMI she+word'] = df.apply(lambda row: pmi(she_count, female_scientist_u[row.name], \n",
    "                                              row[\"she+word count\"], n_women), axis=1)\n",
    "df['PMI he+word'] = df.apply(lambda row: pmi(he_count, male_scientist_u[row.name], \n",
    "                                             row[\"he+word count\"], n_men), axis=1)\n",
    "df.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Log probabilities of word given he/she. Add 1 for smoothing\n",
    "df['log prob of word | he'] = np.log2(\n",
    "    (df.fillna(0)[\"he+word count\"] + 1)) - np.log2((sum(he.values()) + 1))\n",
    "df['log prob of word | she'] = np.log2(\n",
    "    (df.fillna(0)[\"she+word count\"] + 1)) - np.log2((sum(she.values()) + 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "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>she+word count</th>\n",
       "      <th>he+word count</th>\n",
       "      <th>total count</th>\n",
       "      <th>PMI she+word</th>\n",
       "      <th>PMI he+word</th>\n",
       "      <th>log prob of word | he</th>\n",
       "      <th>log prob of word | she</th>\n",
       "      <th>LLR</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>married</th>\n",
       "      <td>250</td>\n",
       "      <td>568</td>\n",
       "      <td>818</td>\n",
       "      <td>4.784584</td>\n",
       "      <td>4.836546</td>\n",
       "      <td>-7.307499</td>\n",
       "      <td>-6.481727</td>\n",
       "      <td>0.825772</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>traveled</th>\n",
       "      <td>43</td>\n",
       "      <td>103</td>\n",
       "      <td>146</td>\n",
       "      <td>5.441462</td>\n",
       "      <td>5.153327</td>\n",
       "      <td>-9.759344</td>\n",
       "      <td>-8.993839</td>\n",
       "      <td>0.765505</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>wanted</th>\n",
       "      <td>58</td>\n",
       "      <td>146</td>\n",
       "      <td>204</td>\n",
       "      <td>4.628066</td>\n",
       "      <td>4.892384</td>\n",
       "      <td>-9.260111</td>\n",
       "      <td>-8.570628</td>\n",
       "      <td>0.689484</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>told</th>\n",
       "      <td>31</td>\n",
       "      <td>80</td>\n",
       "      <td>111</td>\n",
       "      <td>4.116951</td>\n",
       "      <td>4.027025</td>\n",
       "      <td>-10.119934</td>\n",
       "      <td>-9.453271</td>\n",
       "      <td>0.666663</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>met</th>\n",
       "      <td>154</td>\n",
       "      <td>397</td>\n",
       "      <td>551</td>\n",
       "      <td>5.619019</td>\n",
       "      <td>5.266252</td>\n",
       "      <td>-7.823159</td>\n",
       "      <td>-7.177146</td>\n",
       "      <td>0.646013</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>currently</th>\n",
       "      <td>60</td>\n",
       "      <td>156</td>\n",
       "      <td>216</td>\n",
       "      <td>4.197195</td>\n",
       "      <td>3.509979</td>\n",
       "      <td>-9.165163</td>\n",
       "      <td>-8.522533</td>\n",
       "      <td>0.642630</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>undertook</th>\n",
       "      <td>30</td>\n",
       "      <td>79</td>\n",
       "      <td>109</td>\n",
       "      <td>5.676975</td>\n",
       "      <td>5.522000</td>\n",
       "      <td>-10.137856</td>\n",
       "      <td>-9.499074</td>\n",
       "      <td>0.638781</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>grew</th>\n",
       "      <td>45</td>\n",
       "      <td>118</td>\n",
       "      <td>163</td>\n",
       "      <td>4.596602</td>\n",
       "      <td>4.375543</td>\n",
       "      <td>-9.564966</td>\n",
       "      <td>-8.929709</td>\n",
       "      <td>0.635257</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>conducted</th>\n",
       "      <td>62</td>\n",
       "      <td>162</td>\n",
       "      <td>224</td>\n",
       "      <td>4.891391</td>\n",
       "      <td>4.482167</td>\n",
       "      <td>-9.111056</td>\n",
       "      <td>-8.475991</td>\n",
       "      <td>0.635065</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>earned</th>\n",
       "      <td>233</td>\n",
       "      <td>605</td>\n",
       "      <td>838</td>\n",
       "      <td>5.639918</td>\n",
       "      <td>5.595789</td>\n",
       "      <td>-7.216610</td>\n",
       "      <td>-6.582906</td>\n",
       "      <td>0.633704</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>completed</th>\n",
       "      <td>189</td>\n",
       "      <td>496</td>\n",
       "      <td>685</td>\n",
       "      <td>5.642329</td>\n",
       "      <td>5.365061</td>\n",
       "      <td>-7.502682</td>\n",
       "      <td>-6.883415</td>\n",
       "      <td>0.619267</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pioneered</th>\n",
       "      <td>28</td>\n",
       "      <td>76</td>\n",
       "      <td>104</td>\n",
       "      <td>5.577440</td>\n",
       "      <td>4.826647</td>\n",
       "      <td>-10.192997</td>\n",
       "      <td>-9.595290</td>\n",
       "      <td>0.597708</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>collected</th>\n",
       "      <td>30</td>\n",
       "      <td>86</td>\n",
       "      <td>116</td>\n",
       "      <td>4.267585</td>\n",
       "      <td>3.793269</td>\n",
       "      <td>-10.016840</td>\n",
       "      <td>-9.499074</td>\n",
       "      <td>0.517766</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>began</th>\n",
       "      <td>263</td>\n",
       "      <td>741</td>\n",
       "      <td>1004</td>\n",
       "      <td>5.147226</td>\n",
       "      <td>4.948343</td>\n",
       "      <td>-6.924508</td>\n",
       "      <td>-6.408877</td>\n",
       "      <td>0.515632</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>attended</th>\n",
       "      <td>252</td>\n",
       "      <td>711</td>\n",
       "      <td>963</td>\n",
       "      <td>5.574937</td>\n",
       "      <td>5.554586</td>\n",
       "      <td>-6.984050</td>\n",
       "      <td>-6.470277</td>\n",
       "      <td>0.513773</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>stayed</th>\n",
       "      <td>50</td>\n",
       "      <td>145</td>\n",
       "      <td>195</td>\n",
       "      <td>5.922088</td>\n",
       "      <td>5.600125</td>\n",
       "      <td>-9.269959</td>\n",
       "      <td>-8.780845</td>\n",
       "      <td>0.489114</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>and</th>\n",
       "      <td>250</td>\n",
       "      <td>744</td>\n",
       "      <td>994</td>\n",
       "      <td>-1.109155</td>\n",
       "      <td>-1.648303</td>\n",
       "      <td>-6.918687</td>\n",
       "      <td>-6.481727</td>\n",
       "      <td>0.436960</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>continued</th>\n",
       "      <td>218</td>\n",
       "      <td>662</td>\n",
       "      <td>880</td>\n",
       "      <td>5.424527</td>\n",
       "      <td>5.283750</td>\n",
       "      <td>-7.086919</td>\n",
       "      <td>-6.678484</td>\n",
       "      <td>0.408435</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>authored</th>\n",
       "      <td>35</td>\n",
       "      <td>108</td>\n",
       "      <td>143</td>\n",
       "      <td>3.665748</td>\n",
       "      <td>3.149276</td>\n",
       "      <td>-9.691599</td>\n",
       "      <td>-9.283346</td>\n",
       "      <td>0.408254</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>collaborated</th>\n",
       "      <td>29</td>\n",
       "      <td>91</td>\n",
       "      <td>120</td>\n",
       "      <td>5.218675</td>\n",
       "      <td>5.072620</td>\n",
       "      <td>-9.936222</td>\n",
       "      <td>-9.546380</td>\n",
       "      <td>0.389842</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>describes</th>\n",
       "      <td>26</td>\n",
       "      <td>82</td>\n",
       "      <td>108</td>\n",
       "      <td>4.600160</td>\n",
       "      <td>4.345117</td>\n",
       "      <td>-10.084744</td>\n",
       "      <td>-9.698383</td>\n",
       "      <td>0.386361</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>eventually</th>\n",
       "      <td>31</td>\n",
       "      <td>98</td>\n",
       "      <td>129</td>\n",
       "      <td>3.752738</td>\n",
       "      <td>3.433380</td>\n",
       "      <td>-9.830427</td>\n",
       "      <td>-9.453271</td>\n",
       "      <td>0.377157</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>said</th>\n",
       "      <td>123</td>\n",
       "      <td>383</td>\n",
       "      <td>506</td>\n",
       "      <td>4.452856</td>\n",
       "      <td>4.237599</td>\n",
       "      <td>-7.874821</td>\n",
       "      <td>-7.499074</td>\n",
       "      <td>0.375747</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>never</th>\n",
       "      <td>58</td>\n",
       "      <td>182</td>\n",
       "      <td>240</td>\n",
       "      <td>4.116313</td>\n",
       "      <td>3.777676</td>\n",
       "      <td>-8.944084</td>\n",
       "      <td>-8.570628</td>\n",
       "      <td>0.373456</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>applied</th>\n",
       "      <td>35</td>\n",
       "      <td>111</td>\n",
       "      <td>146</td>\n",
       "      <td>3.367986</td>\n",
       "      <td>2.471217</td>\n",
       "      <td>-9.652429</td>\n",
       "      <td>-9.283346</td>\n",
       "      <td>0.369083</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>retired</th>\n",
       "      <td>128</td>\n",
       "      <td>417</td>\n",
       "      <td>545</td>\n",
       "      <td>5.546083</td>\n",
       "      <td>5.671527</td>\n",
       "      <td>-7.752425</td>\n",
       "      <td>-7.442043</td>\n",
       "      <td>0.310381</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gained</th>\n",
       "      <td>43</td>\n",
       "      <td>144</td>\n",
       "      <td>187</td>\n",
       "      <td>5.289459</td>\n",
       "      <td>4.950627</td>\n",
       "      <td>-9.279875</td>\n",
       "      <td>-8.993839</td>\n",
       "      <td>0.286036</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>delivered</th>\n",
       "      <td>23</td>\n",
       "      <td>79</td>\n",
       "      <td>102</td>\n",
       "      <td>4.770085</td>\n",
       "      <td>4.465138</td>\n",
       "      <td>-10.137856</td>\n",
       "      <td>-9.868308</td>\n",
       "      <td>0.269548</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>took</th>\n",
       "      <td>205</td>\n",
       "      <td>696</td>\n",
       "      <td>901</td>\n",
       "      <td>5.196900</td>\n",
       "      <td>5.000298</td>\n",
       "      <td>-7.014769</td>\n",
       "      <td>-6.766770</td>\n",
       "      <td>0.247999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>lived</th>\n",
       "      <td>66</td>\n",
       "      <td>228</td>\n",
       "      <td>294</td>\n",
       "      <td>4.639553</td>\n",
       "      <td>4.751096</td>\n",
       "      <td>-8.620580</td>\n",
       "      <td>-8.387181</td>\n",
       "      <td>0.233399</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              she+word count  he+word count  total count  PMI she+word  \\\n",
       "married                  250            568          818      4.784584   \n",
       "traveled                  43            103          146      5.441462   \n",
       "wanted                    58            146          204      4.628066   \n",
       "told                      31             80          111      4.116951   \n",
       "met                      154            397          551      5.619019   \n",
       "currently                 60            156          216      4.197195   \n",
       "undertook                 30             79          109      5.676975   \n",
       "grew                      45            118          163      4.596602   \n",
       "conducted                 62            162          224      4.891391   \n",
       "earned                   233            605          838      5.639918   \n",
       "completed                189            496          685      5.642329   \n",
       "pioneered                 28             76          104      5.577440   \n",
       "collected                 30             86          116      4.267585   \n",
       "began                    263            741         1004      5.147226   \n",
       "attended                 252            711          963      5.574937   \n",
       "stayed                    50            145          195      5.922088   \n",
       "and                      250            744          994     -1.109155   \n",
       "continued                218            662          880      5.424527   \n",
       "authored                  35            108          143      3.665748   \n",
       "collaborated              29             91          120      5.218675   \n",
       "describes                 26             82          108      4.600160   \n",
       "eventually                31             98          129      3.752738   \n",
       "said                     123            383          506      4.452856   \n",
       "never                     58            182          240      4.116313   \n",
       "applied                   35            111          146      3.367986   \n",
       "retired                  128            417          545      5.546083   \n",
       "gained                    43            144          187      5.289459   \n",
       "delivered                 23             79          102      4.770085   \n",
       "took                     205            696          901      5.196900   \n",
       "lived                     66            228          294      4.639553   \n",
       "\n",
       "              PMI he+word  log prob of word | he  log prob of word | she  \\\n",
       "married          4.836546              -7.307499               -6.481727   \n",
       "traveled         5.153327              -9.759344               -8.993839   \n",
       "wanted           4.892384              -9.260111               -8.570628   \n",
       "told             4.027025             -10.119934               -9.453271   \n",
       "met              5.266252              -7.823159               -7.177146   \n",
       "currently        3.509979              -9.165163               -8.522533   \n",
       "undertook        5.522000             -10.137856               -9.499074   \n",
       "grew             4.375543              -9.564966               -8.929709   \n",
       "conducted        4.482167              -9.111056               -8.475991   \n",
       "earned           5.595789              -7.216610               -6.582906   \n",
       "completed        5.365061              -7.502682               -6.883415   \n",
       "pioneered        4.826647             -10.192997               -9.595290   \n",
       "collected        3.793269             -10.016840               -9.499074   \n",
       "began            4.948343              -6.924508               -6.408877   \n",
       "attended         5.554586              -6.984050               -6.470277   \n",
       "stayed           5.600125              -9.269959               -8.780845   \n",
       "and             -1.648303              -6.918687               -6.481727   \n",
       "continued        5.283750              -7.086919               -6.678484   \n",
       "authored         3.149276              -9.691599               -9.283346   \n",
       "collaborated     5.072620              -9.936222               -9.546380   \n",
       "describes        4.345117             -10.084744               -9.698383   \n",
       "eventually       3.433380              -9.830427               -9.453271   \n",
       "said             4.237599              -7.874821               -7.499074   \n",
       "never            3.777676              -8.944084               -8.570628   \n",
       "applied          2.471217              -9.652429               -9.283346   \n",
       "retired          5.671527              -7.752425               -7.442043   \n",
       "gained           4.950627              -9.279875               -8.993839   \n",
       "delivered        4.465138             -10.137856               -9.868308   \n",
       "took             5.000298              -7.014769               -6.766770   \n",
       "lived            4.751096              -8.620580               -8.387181   \n",
       "\n",
       "                   LLR  \n",
       "married       0.825772  \n",
       "traveled      0.765505  \n",
       "wanted        0.689484  \n",
       "told          0.666663  \n",
       "met           0.646013  \n",
       "currently     0.642630  \n",
       "undertook     0.638781  \n",
       "grew          0.635257  \n",
       "conducted     0.635065  \n",
       "earned        0.633704  \n",
       "completed     0.619267  \n",
       "pioneered     0.597708  \n",
       "collected     0.517766  \n",
       "began         0.515632  \n",
       "attended      0.513773  \n",
       "stayed        0.489114  \n",
       "and           0.436960  \n",
       "continued     0.408435  \n",
       "authored      0.408254  \n",
       "collaborated  0.389842  \n",
       "describes     0.386361  \n",
       "eventually    0.377157  \n",
       "said          0.375747  \n",
       "never         0.373456  \n",
       "applied       0.369083  \n",
       "retired       0.310381  \n",
       "gained        0.286036  \n",
       "delivered     0.269548  \n",
       "took          0.247999  \n",
       "lived         0.233399  "
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Log likelihood ratio she : he\n",
    "df['LLR'] = df['log prob of word | she'] - df['log prob of word | he']\n",
    "df.sort_values(by=['LLR'], ascending=False, inplace=True)\n",
    "df.head(30)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "scrolled": true
   },
   "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>she+word count</th>\n",
       "      <th>he+word count</th>\n",
       "      <th>total count</th>\n",
       "      <th>PMI she+word</th>\n",
       "      <th>PMI he+word</th>\n",
       "      <th>log prob of word | he</th>\n",
       "      <th>log prob of word | she</th>\n",
       "      <th>LLR</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>argued</th>\n",
       "      <td>40</td>\n",
       "      <td>243</td>\n",
       "      <td>283</td>\n",
       "      <td>4.912104</td>\n",
       "      <td>4.716261</td>\n",
       "      <td>-8.529046</td>\n",
       "      <td>-9.095719</td>\n",
       "      <td>-0.566672</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>does</th>\n",
       "      <td>13</td>\n",
       "      <td>85</td>\n",
       "      <td>98</td>\n",
       "      <td>3.027581</td>\n",
       "      <td>3.240481</td>\n",
       "      <td>-10.033519</td>\n",
       "      <td>-10.645916</td>\n",
       "      <td>-0.612397</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>believed</th>\n",
       "      <td>45</td>\n",
       "      <td>286</td>\n",
       "      <td>331</td>\n",
       "      <td>4.931021</td>\n",
       "      <td>4.973056</td>\n",
       "      <td>-8.294877</td>\n",
       "      <td>-8.929709</td>\n",
       "      <td>-0.634832</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>soon</th>\n",
       "      <td>15</td>\n",
       "      <td>100</td>\n",
       "      <td>115</td>\n",
       "      <td>3.225764</td>\n",
       "      <td>3.408588</td>\n",
       "      <td>-9.801572</td>\n",
       "      <td>-10.453271</td>\n",
       "      <td>-0.651698</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>succeeded</th>\n",
       "      <td>15</td>\n",
       "      <td>100</td>\n",
       "      <td>115</td>\n",
       "      <td>4.746238</td>\n",
       "      <td>4.880315</td>\n",
       "      <td>-9.801572</td>\n",
       "      <td>-10.453271</td>\n",
       "      <td>-0.651698</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>claimed</th>\n",
       "      <td>19</td>\n",
       "      <td>126</td>\n",
       "      <td>145</td>\n",
       "      <td>4.210657</td>\n",
       "      <td>4.380913</td>\n",
       "      <td>-9.471099</td>\n",
       "      <td>-10.131343</td>\n",
       "      <td>-0.660243</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>tried</th>\n",
       "      <td>14</td>\n",
       "      <td>95</td>\n",
       "      <td>109</td>\n",
       "      <td>4.511351</td>\n",
       "      <td>4.539003</td>\n",
       "      <td>-9.874821</td>\n",
       "      <td>-10.546380</td>\n",
       "      <td>-0.671559</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>thought</th>\n",
       "      <td>18</td>\n",
       "      <td>125</td>\n",
       "      <td>143</td>\n",
       "      <td>2.841978</td>\n",
       "      <td>2.874453</td>\n",
       "      <td>-9.482504</td>\n",
       "      <td>-10.205343</td>\n",
       "      <td>-0.722839</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>finished</th>\n",
       "      <td>16</td>\n",
       "      <td>113</td>\n",
       "      <td>129</td>\n",
       "      <td>5.155375</td>\n",
       "      <td>5.244570</td>\n",
       "      <td>-9.626894</td>\n",
       "      <td>-10.365808</td>\n",
       "      <td>-0.738914</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>performed</th>\n",
       "      <td>13</td>\n",
       "      <td>93</td>\n",
       "      <td>106</td>\n",
       "      <td>3.784024</td>\n",
       "      <td>4.319861</td>\n",
       "      <td>-9.905195</td>\n",
       "      <td>-10.645916</td>\n",
       "      <td>-0.740721</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>shared</th>\n",
       "      <td>15</td>\n",
       "      <td>108</td>\n",
       "      <td>123</td>\n",
       "      <td>3.547692</td>\n",
       "      <td>4.241225</td>\n",
       "      <td>-9.691599</td>\n",
       "      <td>-10.453271</td>\n",
       "      <td>-0.761671</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>led</th>\n",
       "      <td>44</td>\n",
       "      <td>314</td>\n",
       "      <td>358</td>\n",
       "      <td>3.027393</td>\n",
       "      <td>3.653110</td>\n",
       "      <td>-8.160576</td>\n",
       "      <td>-8.961418</td>\n",
       "      <td>-0.800842</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>subsequently</th>\n",
       "      <td>18</td>\n",
       "      <td>140</td>\n",
       "      <td>158</td>\n",
       "      <td>3.556306</td>\n",
       "      <td>4.048962</td>\n",
       "      <td>-9.320232</td>\n",
       "      <td>-10.205343</td>\n",
       "      <td>-0.885111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>initiated</th>\n",
       "      <td>10</td>\n",
       "      <td>84</td>\n",
       "      <td>94</td>\n",
       "      <td>4.209370</td>\n",
       "      <td>4.592525</td>\n",
       "      <td>-10.050393</td>\n",
       "      <td>-10.993839</td>\n",
       "      <td>-0.943446</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>suggested</th>\n",
       "      <td>10</td>\n",
       "      <td>85</td>\n",
       "      <td>95</td>\n",
       "      <td>2.973072</td>\n",
       "      <td>3.730905</td>\n",
       "      <td>-10.033519</td>\n",
       "      <td>-10.993839</td>\n",
       "      <td>-0.960320</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>headed</th>\n",
       "      <td>14</td>\n",
       "      <td>117</td>\n",
       "      <td>131</td>\n",
       "      <td>4.348621</td>\n",
       "      <td>4.886005</td>\n",
       "      <td>-9.577141</td>\n",
       "      <td>-10.546380</td>\n",
       "      <td>-0.969239</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>defended</th>\n",
       "      <td>9</td>\n",
       "      <td>81</td>\n",
       "      <td>90</td>\n",
       "      <td>5.185122</td>\n",
       "      <td>5.024769</td>\n",
       "      <td>-10.102232</td>\n",
       "      <td>-11.131343</td>\n",
       "      <td>-1.029111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>considered</th>\n",
       "      <td>18</td>\n",
       "      <td>157</td>\n",
       "      <td>175</td>\n",
       "      <td>2.785192</td>\n",
       "      <td>3.485017</td>\n",
       "      <td>-9.156003</td>\n",
       "      <td>-10.205343</td>\n",
       "      <td>-1.049340</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>designed</th>\n",
       "      <td>15</td>\n",
       "      <td>140</td>\n",
       "      <td>155</td>\n",
       "      <td>3.782158</td>\n",
       "      <td>4.092343</td>\n",
       "      <td>-9.320232</td>\n",
       "      <td>-10.453271</td>\n",
       "      <td>-1.133038</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>called</th>\n",
       "      <td>34</td>\n",
       "      <td>334</td>\n",
       "      <td>368</td>\n",
       "      <td>2.504401</td>\n",
       "      <td>3.532916</td>\n",
       "      <td>-8.071767</td>\n",
       "      <td>-9.323988</td>\n",
       "      <td>-1.252221</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>built</th>\n",
       "      <td>12</td>\n",
       "      <td>127</td>\n",
       "      <td>139</td>\n",
       "      <td>3.332680</td>\n",
       "      <td>3.874862</td>\n",
       "      <td>-9.459784</td>\n",
       "      <td>-10.752831</td>\n",
       "      <td>-1.293047</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>introduced</th>\n",
       "      <td>21</td>\n",
       "      <td>219</td>\n",
       "      <td>240</td>\n",
       "      <td>3.736138</td>\n",
       "      <td>4.552815</td>\n",
       "      <td>-8.678424</td>\n",
       "      <td>-9.993839</td>\n",
       "      <td>-1.315415</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>proposed</th>\n",
       "      <td>17</td>\n",
       "      <td>213</td>\n",
       "      <td>230</td>\n",
       "      <td>3.974905</td>\n",
       "      <td>4.302996</td>\n",
       "      <td>-8.718317</td>\n",
       "      <td>-10.283346</td>\n",
       "      <td>-1.565029</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>coined</th>\n",
       "      <td>4</td>\n",
       "      <td>84</td>\n",
       "      <td>88</td>\n",
       "      <td>3.640802</td>\n",
       "      <td>5.072236</td>\n",
       "      <td>-10.050393</td>\n",
       "      <td>-12.131343</td>\n",
       "      <td>-2.080950</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>invented</th>\n",
       "      <td>8</td>\n",
       "      <td>185</td>\n",
       "      <td>193</td>\n",
       "      <td>3.937195</td>\n",
       "      <td>4.710268</td>\n",
       "      <td>-8.920625</td>\n",
       "      <td>-11.283346</td>\n",
       "      <td>-2.362721</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              she+word count  he+word count  total count  PMI she+word  \\\n",
       "argued                    40            243          283      4.912104   \n",
       "does                      13             85           98      3.027581   \n",
       "believed                  45            286          331      4.931021   \n",
       "soon                      15            100          115      3.225764   \n",
       "succeeded                 15            100          115      4.746238   \n",
       "claimed                   19            126          145      4.210657   \n",
       "tried                     14             95          109      4.511351   \n",
       "thought                   18            125          143      2.841978   \n",
       "finished                  16            113          129      5.155375   \n",
       "performed                 13             93          106      3.784024   \n",
       "shared                    15            108          123      3.547692   \n",
       "led                       44            314          358      3.027393   \n",
       "subsequently              18            140          158      3.556306   \n",
       "initiated                 10             84           94      4.209370   \n",
       "suggested                 10             85           95      2.973072   \n",
       "headed                    14            117          131      4.348621   \n",
       "defended                   9             81           90      5.185122   \n",
       "considered                18            157          175      2.785192   \n",
       "designed                  15            140          155      3.782158   \n",
       "called                    34            334          368      2.504401   \n",
       "built                     12            127          139      3.332680   \n",
       "introduced                21            219          240      3.736138   \n",
       "proposed                  17            213          230      3.974905   \n",
       "coined                     4             84           88      3.640802   \n",
       "invented                   8            185          193      3.937195   \n",
       "\n",
       "              PMI he+word  log prob of word | he  log prob of word | she  \\\n",
       "argued           4.716261              -8.529046               -9.095719   \n",
       "does             3.240481             -10.033519              -10.645916   \n",
       "believed         4.973056              -8.294877               -8.929709   \n",
       "soon             3.408588              -9.801572              -10.453271   \n",
       "succeeded        4.880315              -9.801572              -10.453271   \n",
       "claimed          4.380913              -9.471099              -10.131343   \n",
       "tried            4.539003              -9.874821              -10.546380   \n",
       "thought          2.874453              -9.482504              -10.205343   \n",
       "finished         5.244570              -9.626894              -10.365808   \n",
       "performed        4.319861              -9.905195              -10.645916   \n",
       "shared           4.241225              -9.691599              -10.453271   \n",
       "led              3.653110              -8.160576               -8.961418   \n",
       "subsequently     4.048962              -9.320232              -10.205343   \n",
       "initiated        4.592525             -10.050393              -10.993839   \n",
       "suggested        3.730905             -10.033519              -10.993839   \n",
       "headed           4.886005              -9.577141              -10.546380   \n",
       "defended         5.024769             -10.102232              -11.131343   \n",
       "considered       3.485017              -9.156003              -10.205343   \n",
       "designed         4.092343              -9.320232              -10.453271   \n",
       "called           3.532916              -8.071767               -9.323988   \n",
       "built            3.874862              -9.459784              -10.752831   \n",
       "introduced       4.552815              -8.678424               -9.993839   \n",
       "proposed         4.302996              -8.718317              -10.283346   \n",
       "coined           5.072236             -10.050393              -12.131343   \n",
       "invented         4.710268              -8.920625              -11.283346   \n",
       "\n",
       "                   LLR  \n",
       "argued       -0.566672  \n",
       "does         -0.612397  \n",
       "believed     -0.634832  \n",
       "soon         -0.651698  \n",
       "succeeded    -0.651698  \n",
       "claimed      -0.660243  \n",
       "tried        -0.671559  \n",
       "thought      -0.722839  \n",
       "finished     -0.738914  \n",
       "performed    -0.740721  \n",
       "shared       -0.761671  \n",
       "led          -0.800842  \n",
       "subsequently -0.885111  \n",
       "initiated    -0.943446  \n",
       "suggested    -0.960320  \n",
       "headed       -0.969239  \n",
       "defended     -1.029111  \n",
       "considered   -1.049340  \n",
       "designed     -1.133038  \n",
       "called       -1.252221  \n",
       "built        -1.293047  \n",
       "introduced   -1.315415  \n",
       "proposed     -1.565029  \n",
       "coined       -2.080950  \n",
       "invented     -2.362721  "
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.tail(25)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "he_words = df.drop(df[df['PMI he+word'] <= 0 ].index).tail(100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "she_words = df.drop(df[df['PMI she+word'] <= 0 ].index).head(100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "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>she+word count</th>\n",
       "      <th>he+word count</th>\n",
       "      <th>total count</th>\n",
       "      <th>PMI she+word</th>\n",
       "      <th>PMI he+word</th>\n",
       "      <th>log prob of word | he</th>\n",
       "      <th>log prob of word | she</th>\n",
       "      <th>LLR</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>claimed</th>\n",
       "      <td>19</td>\n",
       "      <td>126</td>\n",
       "      <td>145</td>\n",
       "      <td>4.210657</td>\n",
       "      <td>4.380913</td>\n",
       "      <td>-9.471099</td>\n",
       "      <td>-10.131343</td>\n",
       "      <td>-0.660243</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>tried</th>\n",
       "      <td>14</td>\n",
       "      <td>95</td>\n",
       "      <td>109</td>\n",
       "      <td>4.511351</td>\n",
       "      <td>4.539003</td>\n",
       "      <td>-9.874821</td>\n",
       "      <td>-10.546380</td>\n",
       "      <td>-0.671559</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>thought</th>\n",
       "      <td>18</td>\n",
       "      <td>125</td>\n",
       "      <td>143</td>\n",
       "      <td>2.841978</td>\n",
       "      <td>2.874453</td>\n",
       "      <td>-9.482504</td>\n",
       "      <td>-10.205343</td>\n",
       "      <td>-0.722839</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>finished</th>\n",
       "      <td>16</td>\n",
       "      <td>113</td>\n",
       "      <td>129</td>\n",
       "      <td>5.155375</td>\n",
       "      <td>5.244570</td>\n",
       "      <td>-9.626894</td>\n",
       "      <td>-10.365808</td>\n",
       "      <td>-0.738914</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>performed</th>\n",
       "      <td>13</td>\n",
       "      <td>93</td>\n",
       "      <td>106</td>\n",
       "      <td>3.784024</td>\n",
       "      <td>4.319861</td>\n",
       "      <td>-9.905195</td>\n",
       "      <td>-10.645916</td>\n",
       "      <td>-0.740721</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>shared</th>\n",
       "      <td>15</td>\n",
       "      <td>108</td>\n",
       "      <td>123</td>\n",
       "      <td>3.547692</td>\n",
       "      <td>4.241225</td>\n",
       "      <td>-9.691599</td>\n",
       "      <td>-10.453271</td>\n",
       "      <td>-0.761671</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>led</th>\n",
       "      <td>44</td>\n",
       "      <td>314</td>\n",
       "      <td>358</td>\n",
       "      <td>3.027393</td>\n",
       "      <td>3.653110</td>\n",
       "      <td>-8.160576</td>\n",
       "      <td>-8.961418</td>\n",
       "      <td>-0.800842</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>subsequently</th>\n",
       "      <td>18</td>\n",
       "      <td>140</td>\n",
       "      <td>158</td>\n",
       "      <td>3.556306</td>\n",
       "      <td>4.048962</td>\n",
       "      <td>-9.320232</td>\n",
       "      <td>-10.205343</td>\n",
       "      <td>-0.885111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>initiated</th>\n",
       "      <td>10</td>\n",
       "      <td>84</td>\n",
       "      <td>94</td>\n",
       "      <td>4.209370</td>\n",
       "      <td>4.592525</td>\n",
       "      <td>-10.050393</td>\n",
       "      <td>-10.993839</td>\n",
       "      <td>-0.943446</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>suggested</th>\n",
       "      <td>10</td>\n",
       "      <td>85</td>\n",
       "      <td>95</td>\n",
       "      <td>2.973072</td>\n",
       "      <td>3.730905</td>\n",
       "      <td>-10.033519</td>\n",
       "      <td>-10.993839</td>\n",
       "      <td>-0.960320</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>headed</th>\n",
       "      <td>14</td>\n",
       "      <td>117</td>\n",
       "      <td>131</td>\n",
       "      <td>4.348621</td>\n",
       "      <td>4.886005</td>\n",
       "      <td>-9.577141</td>\n",
       "      <td>-10.546380</td>\n",
       "      <td>-0.969239</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>defended</th>\n",
       "      <td>9</td>\n",
       "      <td>81</td>\n",
       "      <td>90</td>\n",
       "      <td>5.185122</td>\n",
       "      <td>5.024769</td>\n",
       "      <td>-10.102232</td>\n",
       "      <td>-11.131343</td>\n",
       "      <td>-1.029111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>considered</th>\n",
       "      <td>18</td>\n",
       "      <td>157</td>\n",
       "      <td>175</td>\n",
       "      <td>2.785192</td>\n",
       "      <td>3.485017</td>\n",
       "      <td>-9.156003</td>\n",
       "      <td>-10.205343</td>\n",
       "      <td>-1.049340</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>designed</th>\n",
       "      <td>15</td>\n",
       "      <td>140</td>\n",
       "      <td>155</td>\n",
       "      <td>3.782158</td>\n",
       "      <td>4.092343</td>\n",
       "      <td>-9.320232</td>\n",
       "      <td>-10.453271</td>\n",
       "      <td>-1.133038</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>called</th>\n",
       "      <td>34</td>\n",
       "      <td>334</td>\n",
       "      <td>368</td>\n",
       "      <td>2.504401</td>\n",
       "      <td>3.532916</td>\n",
       "      <td>-8.071767</td>\n",
       "      <td>-9.323988</td>\n",
       "      <td>-1.252221</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>built</th>\n",
       "      <td>12</td>\n",
       "      <td>127</td>\n",
       "      <td>139</td>\n",
       "      <td>3.332680</td>\n",
       "      <td>3.874862</td>\n",
       "      <td>-9.459784</td>\n",
       "      <td>-10.752831</td>\n",
       "      <td>-1.293047</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>introduced</th>\n",
       "      <td>21</td>\n",
       "      <td>219</td>\n",
       "      <td>240</td>\n",
       "      <td>3.736138</td>\n",
       "      <td>4.552815</td>\n",
       "      <td>-8.678424</td>\n",
       "      <td>-9.993839</td>\n",
       "      <td>-1.315415</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>proposed</th>\n",
       "      <td>17</td>\n",
       "      <td>213</td>\n",
       "      <td>230</td>\n",
       "      <td>3.974905</td>\n",
       "      <td>4.302996</td>\n",
       "      <td>-8.718317</td>\n",
       "      <td>-10.283346</td>\n",
       "      <td>-1.565029</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>coined</th>\n",
       "      <td>4</td>\n",
       "      <td>84</td>\n",
       "      <td>88</td>\n",
       "      <td>3.640802</td>\n",
       "      <td>5.072236</td>\n",
       "      <td>-10.050393</td>\n",
       "      <td>-12.131343</td>\n",
       "      <td>-2.080950</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>invented</th>\n",
       "      <td>8</td>\n",
       "      <td>185</td>\n",
       "      <td>193</td>\n",
       "      <td>3.937195</td>\n",
       "      <td>4.710268</td>\n",
       "      <td>-8.920625</td>\n",
       "      <td>-11.283346</td>\n",
       "      <td>-2.362721</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              she+word count  he+word count  total count  PMI she+word  \\\n",
       "claimed                   19            126          145      4.210657   \n",
       "tried                     14             95          109      4.511351   \n",
       "thought                   18            125          143      2.841978   \n",
       "finished                  16            113          129      5.155375   \n",
       "performed                 13             93          106      3.784024   \n",
       "shared                    15            108          123      3.547692   \n",
       "led                       44            314          358      3.027393   \n",
       "subsequently              18            140          158      3.556306   \n",
       "initiated                 10             84           94      4.209370   \n",
       "suggested                 10             85           95      2.973072   \n",
       "headed                    14            117          131      4.348621   \n",
       "defended                   9             81           90      5.185122   \n",
       "considered                18            157          175      2.785192   \n",
       "designed                  15            140          155      3.782158   \n",
       "called                    34            334          368      2.504401   \n",
       "built                     12            127          139      3.332680   \n",
       "introduced                21            219          240      3.736138   \n",
       "proposed                  17            213          230      3.974905   \n",
       "coined                     4             84           88      3.640802   \n",
       "invented                   8            185          193      3.937195   \n",
       "\n",
       "              PMI he+word  log prob of word | he  log prob of word | she  \\\n",
       "claimed          4.380913              -9.471099              -10.131343   \n",
       "tried            4.539003              -9.874821              -10.546380   \n",
       "thought          2.874453              -9.482504              -10.205343   \n",
       "finished         5.244570              -9.626894              -10.365808   \n",
       "performed        4.319861              -9.905195              -10.645916   \n",
       "shared           4.241225              -9.691599              -10.453271   \n",
       "led              3.653110              -8.160576               -8.961418   \n",
       "subsequently     4.048962              -9.320232              -10.205343   \n",
       "initiated        4.592525             -10.050393              -10.993839   \n",
       "suggested        3.730905             -10.033519              -10.993839   \n",
       "headed           4.886005              -9.577141              -10.546380   \n",
       "defended         5.024769             -10.102232              -11.131343   \n",
       "considered       3.485017              -9.156003              -10.205343   \n",
       "designed         4.092343              -9.320232              -10.453271   \n",
       "called           3.532916              -8.071767               -9.323988   \n",
       "built            3.874862              -9.459784              -10.752831   \n",
       "introduced       4.552815              -8.678424               -9.993839   \n",
       "proposed         4.302996              -8.718317              -10.283346   \n",
       "coined           5.072236             -10.050393              -12.131343   \n",
       "invented         4.710268              -8.920625              -11.283346   \n",
       "\n",
       "                   LLR  \n",
       "claimed      -0.660243  \n",
       "tried        -0.671559  \n",
       "thought      -0.722839  \n",
       "finished     -0.738914  \n",
       "performed    -0.740721  \n",
       "shared       -0.761671  \n",
       "led          -0.800842  \n",
       "subsequently -0.885111  \n",
       "initiated    -0.943446  \n",
       "suggested    -0.960320  \n",
       "headed       -0.969239  \n",
       "defended     -1.029111  \n",
       "considered   -1.049340  \n",
       "designed     -1.133038  \n",
       "called       -1.252221  \n",
       "built        -1.293047  \n",
       "introduced   -1.315415  \n",
       "proposed     -1.565029  \n",
       "coined       -2.080950  \n",
       "invented     -2.362721  "
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "he_words.tail(20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "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>she+word count</th>\n",
       "      <th>he+word count</th>\n",
       "      <th>total count</th>\n",
       "      <th>PMI she+word</th>\n",
       "      <th>PMI he+word</th>\n",
       "      <th>log prob of word | he</th>\n",
       "      <th>log prob of word | she</th>\n",
       "      <th>LLR</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>married</th>\n",
       "      <td>250</td>\n",
       "      <td>568</td>\n",
       "      <td>818</td>\n",
       "      <td>4.784584</td>\n",
       "      <td>4.836546</td>\n",
       "      <td>-7.307499</td>\n",
       "      <td>-6.481727</td>\n",
       "      <td>0.825772</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>traveled</th>\n",
       "      <td>43</td>\n",
       "      <td>103</td>\n",
       "      <td>146</td>\n",
       "      <td>5.441462</td>\n",
       "      <td>5.153327</td>\n",
       "      <td>-9.759344</td>\n",
       "      <td>-8.993839</td>\n",
       "      <td>0.765505</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>wanted</th>\n",
       "      <td>58</td>\n",
       "      <td>146</td>\n",
       "      <td>204</td>\n",
       "      <td>4.628066</td>\n",
       "      <td>4.892384</td>\n",
       "      <td>-9.260111</td>\n",
       "      <td>-8.570628</td>\n",
       "      <td>0.689484</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>told</th>\n",
       "      <td>31</td>\n",
       "      <td>80</td>\n",
       "      <td>111</td>\n",
       "      <td>4.116951</td>\n",
       "      <td>4.027025</td>\n",
       "      <td>-10.119934</td>\n",
       "      <td>-9.453271</td>\n",
       "      <td>0.666663</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>met</th>\n",
       "      <td>154</td>\n",
       "      <td>397</td>\n",
       "      <td>551</td>\n",
       "      <td>5.619019</td>\n",
       "      <td>5.266252</td>\n",
       "      <td>-7.823159</td>\n",
       "      <td>-7.177146</td>\n",
       "      <td>0.646013</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>currently</th>\n",
       "      <td>60</td>\n",
       "      <td>156</td>\n",
       "      <td>216</td>\n",
       "      <td>4.197195</td>\n",
       "      <td>3.509979</td>\n",
       "      <td>-9.165163</td>\n",
       "      <td>-8.522533</td>\n",
       "      <td>0.642630</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>undertook</th>\n",
       "      <td>30</td>\n",
       "      <td>79</td>\n",
       "      <td>109</td>\n",
       "      <td>5.676975</td>\n",
       "      <td>5.522000</td>\n",
       "      <td>-10.137856</td>\n",
       "      <td>-9.499074</td>\n",
       "      <td>0.638781</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>grew</th>\n",
       "      <td>45</td>\n",
       "      <td>118</td>\n",
       "      <td>163</td>\n",
       "      <td>4.596602</td>\n",
       "      <td>4.375543</td>\n",
       "      <td>-9.564966</td>\n",
       "      <td>-8.929709</td>\n",
       "      <td>0.635257</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>conducted</th>\n",
       "      <td>62</td>\n",
       "      <td>162</td>\n",
       "      <td>224</td>\n",
       "      <td>4.891391</td>\n",
       "      <td>4.482167</td>\n",
       "      <td>-9.111056</td>\n",
       "      <td>-8.475991</td>\n",
       "      <td>0.635065</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>earned</th>\n",
       "      <td>233</td>\n",
       "      <td>605</td>\n",
       "      <td>838</td>\n",
       "      <td>5.639918</td>\n",
       "      <td>5.595789</td>\n",
       "      <td>-7.216610</td>\n",
       "      <td>-6.582906</td>\n",
       "      <td>0.633704</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>completed</th>\n",
       "      <td>189</td>\n",
       "      <td>496</td>\n",
       "      <td>685</td>\n",
       "      <td>5.642329</td>\n",
       "      <td>5.365061</td>\n",
       "      <td>-7.502682</td>\n",
       "      <td>-6.883415</td>\n",
       "      <td>0.619267</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pioneered</th>\n",
       "      <td>28</td>\n",
       "      <td>76</td>\n",
       "      <td>104</td>\n",
       "      <td>5.577440</td>\n",
       "      <td>4.826647</td>\n",
       "      <td>-10.192997</td>\n",
       "      <td>-9.595290</td>\n",
       "      <td>0.597708</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>collected</th>\n",
       "      <td>30</td>\n",
       "      <td>86</td>\n",
       "      <td>116</td>\n",
       "      <td>4.267585</td>\n",
       "      <td>3.793269</td>\n",
       "      <td>-10.016840</td>\n",
       "      <td>-9.499074</td>\n",
       "      <td>0.517766</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>began</th>\n",
       "      <td>263</td>\n",
       "      <td>741</td>\n",
       "      <td>1004</td>\n",
       "      <td>5.147226</td>\n",
       "      <td>4.948343</td>\n",
       "      <td>-6.924508</td>\n",
       "      <td>-6.408877</td>\n",
       "      <td>0.515632</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>attended</th>\n",
       "      <td>252</td>\n",
       "      <td>711</td>\n",
       "      <td>963</td>\n",
       "      <td>5.574937</td>\n",
       "      <td>5.554586</td>\n",
       "      <td>-6.984050</td>\n",
       "      <td>-6.470277</td>\n",
       "      <td>0.513773</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>stayed</th>\n",
       "      <td>50</td>\n",
       "      <td>145</td>\n",
       "      <td>195</td>\n",
       "      <td>5.922088</td>\n",
       "      <td>5.600125</td>\n",
       "      <td>-9.269959</td>\n",
       "      <td>-8.780845</td>\n",
       "      <td>0.489114</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>continued</th>\n",
       "      <td>218</td>\n",
       "      <td>662</td>\n",
       "      <td>880</td>\n",
       "      <td>5.424527</td>\n",
       "      <td>5.283750</td>\n",
       "      <td>-7.086919</td>\n",
       "      <td>-6.678484</td>\n",
       "      <td>0.408435</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>authored</th>\n",
       "      <td>35</td>\n",
       "      <td>108</td>\n",
       "      <td>143</td>\n",
       "      <td>3.665748</td>\n",
       "      <td>3.149276</td>\n",
       "      <td>-9.691599</td>\n",
       "      <td>-9.283346</td>\n",
       "      <td>0.408254</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>collaborated</th>\n",
       "      <td>29</td>\n",
       "      <td>91</td>\n",
       "      <td>120</td>\n",
       "      <td>5.218675</td>\n",
       "      <td>5.072620</td>\n",
       "      <td>-9.936222</td>\n",
       "      <td>-9.546380</td>\n",
       "      <td>0.389842</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>describes</th>\n",
       "      <td>26</td>\n",
       "      <td>82</td>\n",
       "      <td>108</td>\n",
       "      <td>4.600160</td>\n",
       "      <td>4.345117</td>\n",
       "      <td>-10.084744</td>\n",
       "      <td>-9.698383</td>\n",
       "      <td>0.386361</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>eventually</th>\n",
       "      <td>31</td>\n",
       "      <td>98</td>\n",
       "      <td>129</td>\n",
       "      <td>3.752738</td>\n",
       "      <td>3.433380</td>\n",
       "      <td>-9.830427</td>\n",
       "      <td>-9.453271</td>\n",
       "      <td>0.377157</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>said</th>\n",
       "      <td>123</td>\n",
       "      <td>383</td>\n",
       "      <td>506</td>\n",
       "      <td>4.452856</td>\n",
       "      <td>4.237599</td>\n",
       "      <td>-7.874821</td>\n",
       "      <td>-7.499074</td>\n",
       "      <td>0.375747</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>never</th>\n",
       "      <td>58</td>\n",
       "      <td>182</td>\n",
       "      <td>240</td>\n",
       "      <td>4.116313</td>\n",
       "      <td>3.777676</td>\n",
       "      <td>-8.944084</td>\n",
       "      <td>-8.570628</td>\n",
       "      <td>0.373456</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>applied</th>\n",
       "      <td>35</td>\n",
       "      <td>111</td>\n",
       "      <td>146</td>\n",
       "      <td>3.367986</td>\n",
       "      <td>2.471217</td>\n",
       "      <td>-9.652429</td>\n",
       "      <td>-9.283346</td>\n",
       "      <td>0.369083</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>retired</th>\n",
       "      <td>128</td>\n",
       "      <td>417</td>\n",
       "      <td>545</td>\n",
       "      <td>5.546083</td>\n",
       "      <td>5.671527</td>\n",
       "      <td>-7.752425</td>\n",
       "      <td>-7.442043</td>\n",
       "      <td>0.310381</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gained</th>\n",
       "      <td>43</td>\n",
       "      <td>144</td>\n",
       "      <td>187</td>\n",
       "      <td>5.289459</td>\n",
       "      <td>4.950627</td>\n",
       "      <td>-9.279875</td>\n",
       "      <td>-8.993839</td>\n",
       "      <td>0.286036</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>delivered</th>\n",
       "      <td>23</td>\n",
       "      <td>79</td>\n",
       "      <td>102</td>\n",
       "      <td>4.770085</td>\n",
       "      <td>4.465138</td>\n",
       "      <td>-10.137856</td>\n",
       "      <td>-9.868308</td>\n",
       "      <td>0.269548</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>took</th>\n",
       "      <td>205</td>\n",
       "      <td>696</td>\n",
       "      <td>901</td>\n",
       "      <td>5.196900</td>\n",
       "      <td>5.000298</td>\n",
       "      <td>-7.014769</td>\n",
       "      <td>-6.766770</td>\n",
       "      <td>0.247999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>lived</th>\n",
       "      <td>66</td>\n",
       "      <td>228</td>\n",
       "      <td>294</td>\n",
       "      <td>4.639553</td>\n",
       "      <td>4.751096</td>\n",
       "      <td>-8.620580</td>\n",
       "      <td>-8.387181</td>\n",
       "      <td>0.233399</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>felt</th>\n",
       "      <td>39</td>\n",
       "      <td>136</td>\n",
       "      <td>175</td>\n",
       "      <td>4.680448</td>\n",
       "      <td>5.174122</td>\n",
       "      <td>-9.361752</td>\n",
       "      <td>-9.131343</td>\n",
       "      <td>0.230409</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              she+word count  he+word count  total count  PMI she+word  \\\n",
       "married                  250            568          818      4.784584   \n",
       "traveled                  43            103          146      5.441462   \n",
       "wanted                    58            146          204      4.628066   \n",
       "told                      31             80          111      4.116951   \n",
       "met                      154            397          551      5.619019   \n",
       "currently                 60            156          216      4.197195   \n",
       "undertook                 30             79          109      5.676975   \n",
       "grew                      45            118          163      4.596602   \n",
       "conducted                 62            162          224      4.891391   \n",
       "earned                   233            605          838      5.639918   \n",
       "completed                189            496          685      5.642329   \n",
       "pioneered                 28             76          104      5.577440   \n",
       "collected                 30             86          116      4.267585   \n",
       "began                    263            741         1004      5.147226   \n",
       "attended                 252            711          963      5.574937   \n",
       "stayed                    50            145          195      5.922088   \n",
       "continued                218            662          880      5.424527   \n",
       "authored                  35            108          143      3.665748   \n",
       "collaborated              29             91          120      5.218675   \n",
       "describes                 26             82          108      4.600160   \n",
       "eventually                31             98          129      3.752738   \n",
       "said                     123            383          506      4.452856   \n",
       "never                     58            182          240      4.116313   \n",
       "applied                   35            111          146      3.367986   \n",
       "retired                  128            417          545      5.546083   \n",
       "gained                    43            144          187      5.289459   \n",
       "delivered                 23             79          102      4.770085   \n",
       "took                     205            696          901      5.196900   \n",
       "lived                     66            228          294      4.639553   \n",
       "felt                      39            136          175      4.680448   \n",
       "\n",
       "              PMI he+word  log prob of word | he  log prob of word | she  \\\n",
       "married          4.836546              -7.307499               -6.481727   \n",
       "traveled         5.153327              -9.759344               -8.993839   \n",
       "wanted           4.892384              -9.260111               -8.570628   \n",
       "told             4.027025             -10.119934               -9.453271   \n",
       "met              5.266252              -7.823159               -7.177146   \n",
       "currently        3.509979              -9.165163               -8.522533   \n",
       "undertook        5.522000             -10.137856               -9.499074   \n",
       "grew             4.375543              -9.564966               -8.929709   \n",
       "conducted        4.482167              -9.111056               -8.475991   \n",
       "earned           5.595789              -7.216610               -6.582906   \n",
       "completed        5.365061              -7.502682               -6.883415   \n",
       "pioneered        4.826647             -10.192997               -9.595290   \n",
       "collected        3.793269             -10.016840               -9.499074   \n",
       "began            4.948343              -6.924508               -6.408877   \n",
       "attended         5.554586              -6.984050               -6.470277   \n",
       "stayed           5.600125              -9.269959               -8.780845   \n",
       "continued        5.283750              -7.086919               -6.678484   \n",
       "authored         3.149276              -9.691599               -9.283346   \n",
       "collaborated     5.072620              -9.936222               -9.546380   \n",
       "describes        4.345117             -10.084744               -9.698383   \n",
       "eventually       3.433380              -9.830427               -9.453271   \n",
       "said             4.237599              -7.874821               -7.499074   \n",
       "never            3.777676              -8.944084               -8.570628   \n",
       "applied          2.471217              -9.652429               -9.283346   \n",
       "retired          5.671527              -7.752425               -7.442043   \n",
       "gained           4.950627              -9.279875               -8.993839   \n",
       "delivered        4.465138             -10.137856               -9.868308   \n",
       "took             5.000298              -7.014769               -6.766770   \n",
       "lived            4.751096              -8.620580               -8.387181   \n",
       "felt             5.174122              -9.361752               -9.131343   \n",
       "\n",
       "                   LLR  \n",
       "married       0.825772  \n",
       "traveled      0.765505  \n",
       "wanted        0.689484  \n",
       "told          0.666663  \n",
       "met           0.646013  \n",
       "currently     0.642630  \n",
       "undertook     0.638781  \n",
       "grew          0.635257  \n",
       "conducted     0.635065  \n",
       "earned        0.633704  \n",
       "completed     0.619267  \n",
       "pioneered     0.597708  \n",
       "collected     0.517766  \n",
       "began         0.515632  \n",
       "attended      0.513773  \n",
       "stayed        0.489114  \n",
       "continued     0.408435  \n",
       "authored      0.408254  \n",
       "collaborated  0.389842  \n",
       "describes     0.386361  \n",
       "eventually    0.377157  \n",
       "said          0.375747  \n",
       "never         0.373456  \n",
       "applied       0.369083  \n",
       "retired       0.310381  \n",
       "gained        0.286036  \n",
       "delivered     0.269548  \n",
       "took          0.247999  \n",
       "lived         0.233399  \n",
       "felt          0.230409  "
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "she_words.head(30)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "married      0.825772\n",
       "met          0.646013\n",
       "earned       0.633704\n",
       "completed    0.619267\n",
       "began        0.515632\n",
       "attended     0.513773\n",
       "continued    0.408435\n",
       "said         0.375747\n",
       "retired      0.310381\n",
       "took         0.247999\n",
       "graduated    0.219472\n",
       "won          0.187753\n",
       "started      0.163950\n",
       "spent        0.153445\n",
       "did          0.149376\n",
       "Name: LLR, dtype: float64"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "she_top = she_words.drop(she_words[she_words['she+word count'] < 75 ].index).head(15)['LLR']\n",
    "Y1, X1 = she_top.index, she_top \n",
    "she_top"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "shared         -0.761671\n",
       "led            -0.800842\n",
       "subsequently   -0.885111\n",
       "initiated      -0.943446\n",
       "suggested      -0.960320\n",
       "headed         -0.969239\n",
       "defended       -1.029111\n",
       "considered     -1.049340\n",
       "designed       -1.133038\n",
       "called         -1.252221\n",
       "built          -1.293047\n",
       "introduced     -1.315415\n",
       "proposed       -1.565029\n",
       "coined         -2.080950\n",
       "invented       -2.362721\n",
       "Name: LLR, dtype: float64"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y2, X2 = he_words.tail(15).index, he_words.tail(15)['LLR']\n",
    "he_words.tail(15)['LLR']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 972x936 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "title=\"Log likelihood ratio that word appears after she/he in articles about female and male scientists, respectively\"\n",
    "subtitle=\"Limited to words appearing at least 75 times after pronoun and with positive word association (PMI) for that gender\"\n",
    "subtitle2=\"\\n\\nHe...                                       She...\"\n",
    "\n",
    "Y = np.arange(30)\n",
    "X_women = np.concatenate((np.zeros(15), np.flip(X1.values, 0)), axis=0)\n",
    "X_men = np.concatenate((np.flip(X2.values, 0), np.zeros(15)), axis=0)\n",
    "\n",
    "plt.figure(figsize=(13.5,13))\n",
    "plt.axes([0.025, 0.025, 0.95, 0.95])\n",
    "plt.barh(Y, X_women, facecolor=\"None\", edgecolor='#2ca79b')\n",
    "plt.barh(Y, X_men, facecolor=\"None\", edgecolor='#a52e3b')\n",
    "\n",
    "for x, y, label in zip(X_women, Y, np.concatenate((np.empty(15, dtype=\"U\") , np.flip(Y1,0)))):\n",
    "    plt.text(x+2, y, label, ha='right', va= 'center', color=\"black\", fontsize=13)\n",
    "for x, y, label in zip(X_men, Y, np.concatenate((np.flip(Y2,0), np.empty(15, dtype=\"U\")))):\n",
    "    plt.text(x-3, y, label, ha='left', va= 'center', color=\"black\", fontsize=13)\n",
    "\n",
    "\n",
    "plt.title(title, fontsize=17, y=1.06)\n",
    "plt.suptitle(subtitle + subtitle2, fontsize=14, y=1.03)\n",
    "plt.axvline(x=0, color=\"black\")\n",
    "plt.ylim(-.5, 29.6)\n",
    "plt.yticks(())\n",
    "plt.xlim(-12, 13.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Category: Artists"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "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>she+word count</th>\n",
       "      <th>he+word count</th>\n",
       "      <th>total count</th>\n",
       "      <th>PMI she+word</th>\n",
       "      <th>PMI he+word</th>\n",
       "      <th>log prob of word | he</th>\n",
       "      <th>log prob of word | she</th>\n",
       "      <th>LLR</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>picked</th>\n",
       "      <td>24</td>\n",
       "      <td>95</td>\n",
       "      <td>119</td>\n",
       "      <td>3.561867</td>\n",
       "      <td>4.020610</td>\n",
       "      <td>-11.366186</td>\n",
       "      <td>-12.138425</td>\n",
       "      <td>-0.772240</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>read</th>\n",
       "      <td>31</td>\n",
       "      <td>122</td>\n",
       "      <td>153</td>\n",
       "      <td>1.858919</td>\n",
       "      <td>2.908634</td>\n",
       "      <td>-11.008634</td>\n",
       "      <td>-11.782282</td>\n",
       "      <td>-0.773648</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>led</th>\n",
       "      <td>68</td>\n",
       "      <td>270</td>\n",
       "      <td>338</td>\n",
       "      <td>2.411068</td>\n",
       "      <td>2.680915</td>\n",
       "      <td>-9.868999</td>\n",
       "      <td>-10.673757</td>\n",
       "      <td>-0.804758</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>d</th>\n",
       "      <td>114</td>\n",
       "      <td>455</td>\n",
       "      <td>569</td>\n",
       "      <td>0.857162</td>\n",
       "      <td>1.530431</td>\n",
       "      <td>-9.118258</td>\n",
       "      <td>-9.936792</td>\n",
       "      <td>-0.818533</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ever</th>\n",
       "      <td>23</td>\n",
       "      <td>95</td>\n",
       "      <td>118</td>\n",
       "      <td>0.538650</td>\n",
       "      <td>1.469715</td>\n",
       "      <td>-11.366186</td>\n",
       "      <td>-12.197319</td>\n",
       "      <td>-0.831134</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>directed</th>\n",
       "      <td>64</td>\n",
       "      <td>260</td>\n",
       "      <td>324</td>\n",
       "      <td>2.293125</td>\n",
       "      <td>3.123923</td>\n",
       "      <td>-9.923242</td>\n",
       "      <td>-10.759914</td>\n",
       "      <td>-0.836672</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>produced</th>\n",
       "      <td>332</td>\n",
       "      <td>1345</td>\n",
       "      <td>1677</td>\n",
       "      <td>3.263215</td>\n",
       "      <td>3.901936</td>\n",
       "      <td>-7.556685</td>\n",
       "      <td>-8.402903</td>\n",
       "      <td>-0.846218</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>carried</th>\n",
       "      <td>20</td>\n",
       "      <td>85</td>\n",
       "      <td>105</td>\n",
       "      <td>3.671435</td>\n",
       "      <td>4.055091</td>\n",
       "      <td>-11.524883</td>\n",
       "      <td>-12.389964</td>\n",
       "      <td>-0.865081</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>quit</th>\n",
       "      <td>29</td>\n",
       "      <td>125</td>\n",
       "      <td>154</td>\n",
       "      <td>4.331120</td>\n",
       "      <td>4.423707</td>\n",
       "      <td>-10.973868</td>\n",
       "      <td>-11.875391</td>\n",
       "      <td>-0.901523</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>owned</th>\n",
       "      <td>23</td>\n",
       "      <td>102</td>\n",
       "      <td>125</td>\n",
       "      <td>2.765656</td>\n",
       "      <td>3.212599</td>\n",
       "      <td>-11.264648</td>\n",
       "      <td>-12.197319</td>\n",
       "      <td>-0.932672</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>served</th>\n",
       "      <td>198</td>\n",
       "      <td>861</td>\n",
       "      <td>1059</td>\n",
       "      <td>4.723984</td>\n",
       "      <td>5.153145</td>\n",
       "      <td>-8.199604</td>\n",
       "      <td>-9.145657</td>\n",
       "      <td>-0.946053</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>built</th>\n",
       "      <td>42</td>\n",
       "      <td>188</td>\n",
       "      <td>230</td>\n",
       "      <td>3.192138</td>\n",
       "      <td>3.663883</td>\n",
       "      <td>-10.388906</td>\n",
       "      <td>-11.356017</td>\n",
       "      <td>-0.967111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bought</th>\n",
       "      <td>52</td>\n",
       "      <td>236</td>\n",
       "      <td>288</td>\n",
       "      <td>3.962074</td>\n",
       "      <td>4.584681</td>\n",
       "      <td>-10.062405</td>\n",
       "      <td>-11.054361</td>\n",
       "      <td>-0.991956</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>acquired</th>\n",
       "      <td>31</td>\n",
       "      <td>143</td>\n",
       "      <td>174</td>\n",
       "      <td>3.454159</td>\n",
       "      <td>4.243296</td>\n",
       "      <td>-10.781223</td>\n",
       "      <td>-11.782282</td>\n",
       "      <td>-1.001059</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>defeated</th>\n",
       "      <td>150</td>\n",
       "      <td>680</td>\n",
       "      <td>830</td>\n",
       "      <td>5.160677</td>\n",
       "      <td>4.770896</td>\n",
       "      <td>-8.539637</td>\n",
       "      <td>-9.543877</td>\n",
       "      <td>-1.004240</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>offered</th>\n",
       "      <td>19</td>\n",
       "      <td>100</td>\n",
       "      <td>119</td>\n",
       "      <td>1.922415</td>\n",
       "      <td>2.984507</td>\n",
       "      <td>-11.292937</td>\n",
       "      <td>-12.460354</td>\n",
       "      <td>-1.167417</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>faced</th>\n",
       "      <td>108</td>\n",
       "      <td>583</td>\n",
       "      <td>691</td>\n",
       "      <td>4.598209</td>\n",
       "      <td>4.011733</td>\n",
       "      <td>-8.761324</td>\n",
       "      <td>-10.014097</td>\n",
       "      <td>-1.252774</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>arranged</th>\n",
       "      <td>19</td>\n",
       "      <td>108</td>\n",
       "      <td>127</td>\n",
       "      <td>2.526050</td>\n",
       "      <td>3.460121</td>\n",
       "      <td>-11.182964</td>\n",
       "      <td>-12.460354</td>\n",
       "      <td>-1.277390</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>heard</th>\n",
       "      <td>43</td>\n",
       "      <td>242</td>\n",
       "      <td>285</td>\n",
       "      <td>2.722745</td>\n",
       "      <td>3.698490</td>\n",
       "      <td>-10.026336</td>\n",
       "      <td>-11.322850</td>\n",
       "      <td>-1.296514</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>beat</th>\n",
       "      <td>19</td>\n",
       "      <td>111</td>\n",
       "      <td>130</td>\n",
       "      <td>1.082687</td>\n",
       "      <td>1.989080</td>\n",
       "      <td>-11.143793</td>\n",
       "      <td>-12.460354</td>\n",
       "      <td>-1.316560</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>executed</th>\n",
       "      <td>16</td>\n",
       "      <td>98</td>\n",
       "      <td>114</td>\n",
       "      <td>3.308285</td>\n",
       "      <td>4.042409</td>\n",
       "      <td>-11.321791</td>\n",
       "      <td>-12.694819</td>\n",
       "      <td>-1.373027</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>formed</th>\n",
       "      <td>133</td>\n",
       "      <td>780</td>\n",
       "      <td>913</td>\n",
       "      <td>4.240296</td>\n",
       "      <td>4.720156</td>\n",
       "      <td>-8.341969</td>\n",
       "      <td>-9.716192</td>\n",
       "      <td>-1.374223</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>drew</th>\n",
       "      <td>92</td>\n",
       "      <td>740</td>\n",
       "      <td>832</td>\n",
       "      <td>3.913711</td>\n",
       "      <td>4.848620</td>\n",
       "      <td>-8.417818</td>\n",
       "      <td>-10.243123</td>\n",
       "      <td>-1.825304</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>conducted</th>\n",
       "      <td>19</td>\n",
       "      <td>159</td>\n",
       "      <td>178</td>\n",
       "      <td>2.890042</td>\n",
       "      <td>4.086907</td>\n",
       "      <td>-10.629220</td>\n",
       "      <td>-12.460354</td>\n",
       "      <td>-1.831134</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>composed</th>\n",
       "      <td>51</td>\n",
       "      <td>425</td>\n",
       "      <td>476</td>\n",
       "      <td>2.769976</td>\n",
       "      <td>4.278643</td>\n",
       "      <td>-9.216438</td>\n",
       "      <td>-11.081842</td>\n",
       "      <td>-1.865403</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>lost</th>\n",
       "      <td>218</td>\n",
       "      <td>2308</td>\n",
       "      <td>2526</td>\n",
       "      <td>3.847560</td>\n",
       "      <td>5.100009</td>\n",
       "      <td>-6.778096</td>\n",
       "      <td>-9.007495</td>\n",
       "      <td>-2.229399</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>enlisted</th>\n",
       "      <td>9</td>\n",
       "      <td>105</td>\n",
       "      <td>114</td>\n",
       "      <td>3.984100</td>\n",
       "      <td>5.114066</td>\n",
       "      <td>-11.223228</td>\n",
       "      <td>-13.460354</td>\n",
       "      <td>-2.237126</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fought</th>\n",
       "      <td>29</td>\n",
       "      <td>503</td>\n",
       "      <td>532</td>\n",
       "      <td>4.221495</td>\n",
       "      <td>4.568853</td>\n",
       "      <td>-8.973868</td>\n",
       "      <td>-11.875391</td>\n",
       "      <td>-2.901523</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>knocked</th>\n",
       "      <td>2</td>\n",
       "      <td>90</td>\n",
       "      <td>92</td>\n",
       "      <td>1.650677</td>\n",
       "      <td>3.758254</td>\n",
       "      <td>-11.443353</td>\n",
       "      <td>-15.197319</td>\n",
       "      <td>-3.753966</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>himself</th>\n",
       "      <td>0</td>\n",
       "      <td>142</td>\n",
       "      <td>142</td>\n",
       "      <td>-100.000000</td>\n",
       "      <td>0.631674</td>\n",
       "      <td>-10.791277</td>\n",
       "      <td>-16.782282</td>\n",
       "      <td>-5.991005</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           she+word count  he+word count  total count  PMI she+word  \\\n",
       "picked                 24             95          119      3.561867   \n",
       "read                   31            122          153      1.858919   \n",
       "led                    68            270          338      2.411068   \n",
       "d                     114            455          569      0.857162   \n",
       "ever                   23             95          118      0.538650   \n",
       "directed               64            260          324      2.293125   \n",
       "produced              332           1345         1677      3.263215   \n",
       "carried                20             85          105      3.671435   \n",
       "quit                   29            125          154      4.331120   \n",
       "owned                  23            102          125      2.765656   \n",
       "served                198            861         1059      4.723984   \n",
       "built                  42            188          230      3.192138   \n",
       "bought                 52            236          288      3.962074   \n",
       "acquired               31            143          174      3.454159   \n",
       "defeated              150            680          830      5.160677   \n",
       "offered                19            100          119      1.922415   \n",
       "faced                 108            583          691      4.598209   \n",
       "arranged               19            108          127      2.526050   \n",
       "heard                  43            242          285      2.722745   \n",
       "beat                   19            111          130      1.082687   \n",
       "executed               16             98          114      3.308285   \n",
       "formed                133            780          913      4.240296   \n",
       "drew                   92            740          832      3.913711   \n",
       "conducted              19            159          178      2.890042   \n",
       "composed               51            425          476      2.769976   \n",
       "lost                  218           2308         2526      3.847560   \n",
       "enlisted                9            105          114      3.984100   \n",
       "fought                 29            503          532      4.221495   \n",
       "knocked                 2             90           92      1.650677   \n",
       "himself                 0            142          142   -100.000000   \n",
       "\n",
       "           PMI he+word  log prob of word | he  log prob of word | she  \\\n",
       "picked        4.020610             -11.366186              -12.138425   \n",
       "read          2.908634             -11.008634              -11.782282   \n",
       "led           2.680915              -9.868999              -10.673757   \n",
       "d             1.530431              -9.118258               -9.936792   \n",
       "ever          1.469715             -11.366186              -12.197319   \n",
       "directed      3.123923              -9.923242              -10.759914   \n",
       "produced      3.901936              -7.556685               -8.402903   \n",
       "carried       4.055091             -11.524883              -12.389964   \n",
       "quit          4.423707             -10.973868              -11.875391   \n",
       "owned         3.212599             -11.264648              -12.197319   \n",
       "served        5.153145              -8.199604               -9.145657   \n",
       "built         3.663883             -10.388906              -11.356017   \n",
       "bought        4.584681             -10.062405              -11.054361   \n",
       "acquired      4.243296             -10.781223              -11.782282   \n",
       "defeated      4.770896              -8.539637               -9.543877   \n",
       "offered       2.984507             -11.292937              -12.460354   \n",
       "faced         4.011733              -8.761324              -10.014097   \n",
       "arranged      3.460121             -11.182964              -12.460354   \n",
       "heard         3.698490             -10.026336              -11.322850   \n",
       "beat          1.989080             -11.143793              -12.460354   \n",
       "executed      4.042409             -11.321791              -12.694819   \n",
       "formed        4.720156              -8.341969               -9.716192   \n",
       "drew          4.848620              -8.417818              -10.243123   \n",
       "conducted     4.086907             -10.629220              -12.460354   \n",
       "composed      4.278643              -9.216438              -11.081842   \n",
       "lost          5.100009              -6.778096               -9.007495   \n",
       "enlisted      5.114066             -11.223228              -13.460354   \n",
       "fought        4.568853              -8.973868              -11.875391   \n",
       "knocked       3.758254             -11.443353              -15.197319   \n",
       "himself       0.631674             -10.791277              -16.782282   \n",
       "\n",
       "                LLR  \n",
       "picked    -0.772240  \n",
       "read      -0.773648  \n",
       "led       -0.804758  \n",
       "d         -0.818533  \n",
       "ever      -0.831134  \n",
       "directed  -0.836672  \n",
       "produced  -0.846218  \n",
       "carried   -0.865081  \n",
       "quit      -0.901523  \n",
       "owned     -0.932672  \n",
       "served    -0.946053  \n",
       "built     -0.967111  \n",
       "bought    -0.991956  \n",
       "acquired  -1.001059  \n",
       "defeated  -1.004240  \n",
       "offered   -1.167417  \n",
       "faced     -1.252774  \n",
       "arranged  -1.277390  \n",
       "heard     -1.296514  \n",
       "beat      -1.316560  \n",
       "executed  -1.373027  \n",
       "formed    -1.374223  \n",
       "drew      -1.825304  \n",
       "conducted -1.831134  \n",
       "composed  -1.865403  \n",
       "lost      -2.229399  \n",
       "enlisted  -2.237126  \n",
       "fought    -2.901523  \n",
       "knocked   -3.753966  \n",
       "himself   -5.991005  "
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Number of words in corpus of men and women's ARTISTS bios: \n",
    "n_women = sum(female_artists_u.values())\n",
    "n_men = sum(male_artists_u.values())\n",
    "\n",
    "# bigrams beginning with he or she, filtered by words appearing at least 75 times.\n",
    "she = collections.Counter({ k[1]: v for k, v in female_artists.items() if v >= 75 and k[0] == \"she\" })\n",
    "he = collections.Counter({ k[1]: v for k, v in male_artists.items() if v >= 75 and k[0] == \"he\" })\n",
    "\n",
    "he_count = male_artists_u['he']\n",
    "she_count = female_artists_u['she']\n",
    "\n",
    "for word in he:\n",
    "    if word not in she:\n",
    "        she[word] = female_artists[(\"she\", word)]\n",
    "for word in she:\n",
    "    if word not in he:\n",
    "        he[word] = male_artists[(\"he\", word)]\n",
    "\n",
    "columns = ['Word following he/she', \"she+word count\"]\n",
    "df = pd.DataFrame([(k,v) for k,v in she.items()], columns=columns)\n",
    "df.set_index(['Word following he/she'], inplace=True)\n",
    "columns2 = ['Word following he/she', \"he+word count\",]\n",
    "df2 = pd.DataFrame([(k,v) for k,v in he.items()], columns=columns2)\n",
    "df2.set_index(['Word following he/she'], inplace=True)\n",
    "\n",
    "df = pd.concat([df,df2], axis=1)\n",
    "df['total count'] = df.fillna(0)[\"she+word count\"] + df.fillna(0)[\"he+word count\"]\n",
    "df.sort_values(by=['total count'], ascending=False, inplace=True)\n",
    "\n",
    "df['PMI she+word'] = df.apply(lambda row: pmi(she_count, female_artists_u[row.name], \n",
    "                                              row[\"she+word count\"], n_women), axis=1)\n",
    "df['PMI he+word'] = df.apply(lambda row: pmi(he_count, male_artists_u[row.name], \n",
    "                                             row[\"he+word count\"], n_men), axis=1)\n",
    "\n",
    "# Log probabilities of word given he/she. Add 1 for smoothing\n",
    "df['log prob of word | he'] = np.log2(\n",
    "    (df.fillna(0)[\"he+word count\"] + 1)) - np.log2((sum(he.values()) + 1))\n",
    "df['log prob of word | she'] = np.log2(\n",
    "    (df.fillna(0)[\"she+word count\"] + 1)) - np.log2((sum(she.values()) + 1))\n",
    "\n",
    "# Log likelihood ratio she : he\n",
    "df['LLR'] = df['log prob of word | she'] - df['log prob of word | he']\n",
    "df.sort_values(by=['LLR'], ascending=False, inplace=True)\n",
    "\n",
    "\n",
    "he_words = df.drop(df[df['PMI he+word'] <= 0 ].index).tail(100)\n",
    "she_words = df.drop(df[df['PMI she+word'] <= 0 ].index).head(100)\n",
    "\n",
    "he_words.tail(30)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "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>she+word count</th>\n",
       "      <th>he+word count</th>\n",
       "      <th>total count</th>\n",
       "      <th>PMI she+word</th>\n",
       "      <th>PMI he+word</th>\n",
       "      <th>log prob of word | he</th>\n",
       "      <th>log prob of word | she</th>\n",
       "      <th>LLR</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>placed</th>\n",
       "      <td>836</td>\n",
       "      <td>155</td>\n",
       "      <td>991</td>\n",
       "      <td>5.852508</td>\n",
       "      <td>3.584022</td>\n",
       "      <td>-10.665746</td>\n",
       "      <td>-7.073198</td>\n",
       "      <td>3.592548</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>qualified</th>\n",
       "      <td>119</td>\n",
       "      <td>61</td>\n",
       "      <td>180</td>\n",
       "      <td>5.265387</td>\n",
       "      <td>4.799512</td>\n",
       "      <td>-11.996952</td>\n",
       "      <td>-9.875391</td>\n",
       "      <td>2.121561</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sang</th>\n",
       "      <td>1025</td>\n",
       "      <td>830</td>\n",
       "      <td>1855</td>\n",
       "      <td>5.552048</td>\n",
       "      <td>5.252919</td>\n",
       "      <td>-8.252443</td>\n",
       "      <td>-6.779467</td>\n",
       "      <td>1.472977</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>auditioned</th>\n",
       "      <td>90</td>\n",
       "      <td>74</td>\n",
       "      <td>164</td>\n",
       "      <td>5.452698</td>\n",
       "      <td>5.192511</td>\n",
       "      <td>-11.722329</td>\n",
       "      <td>-10.274487</td>\n",
       "      <td>1.447842</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sings</th>\n",
       "      <td>128</td>\n",
       "      <td>116</td>\n",
       "      <td>244</td>\n",
       "      <td>3.818992</td>\n",
       "      <td>3.368039</td>\n",
       "      <td>-11.080783</td>\n",
       "      <td>-9.771054</td>\n",
       "      <td>1.309729</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>starred</th>\n",
       "      <td>438</td>\n",
       "      <td>412</td>\n",
       "      <td>850</td>\n",
       "      <td>5.037241</td>\n",
       "      <td>4.752705</td>\n",
       "      <td>-9.261150</td>\n",
       "      <td>-8.004204</td>\n",
       "      <td>1.256946</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>competed</th>\n",
       "      <td>450</td>\n",
       "      <td>426</td>\n",
       "      <td>876</td>\n",
       "      <td>5.256862</td>\n",
       "      <td>4.774455</td>\n",
       "      <td>-9.213056</td>\n",
       "      <td>-7.965298</td>\n",
       "      <td>1.247758</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>photographed</th>\n",
       "      <td>81</td>\n",
       "      <td>85</td>\n",
       "      <td>166</td>\n",
       "      <td>4.371617</td>\n",
       "      <td>4.309556</td>\n",
       "      <td>-11.524883</td>\n",
       "      <td>-10.424730</td>\n",
       "      <td>1.100154</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>scored</th>\n",
       "      <td>181</td>\n",
       "      <td>191</td>\n",
       "      <td>372</td>\n",
       "      <td>5.565560</td>\n",
       "      <td>4.683200</td>\n",
       "      <td>-10.366186</td>\n",
       "      <td>-9.274487</td>\n",
       "      <td>1.091699</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>finished</th>\n",
       "      <td>391</td>\n",
       "      <td>416</td>\n",
       "      <td>807</td>\n",
       "      <td>5.481591</td>\n",
       "      <td>4.766644</td>\n",
       "      <td>-9.247245</td>\n",
       "      <td>-8.167572</td>\n",
       "      <td>1.079673</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>revealed</th>\n",
       "      <td>93</td>\n",
       "      <td>101</td>\n",
       "      <td>194</td>\n",
       "      <td>3.366620</td>\n",
       "      <td>2.599228</td>\n",
       "      <td>-11.278723</td>\n",
       "      <td>-10.227693</td>\n",
       "      <td>1.051030</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>acted</th>\n",
       "      <td>88</td>\n",
       "      <td>100</td>\n",
       "      <td>188</td>\n",
       "      <td>4.940183</td>\n",
       "      <td>4.563683</td>\n",
       "      <td>-11.292937</td>\n",
       "      <td>-10.306548</td>\n",
       "      <td>0.986388</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>wore</th>\n",
       "      <td>73</td>\n",
       "      <td>88</td>\n",
       "      <td>161</td>\n",
       "      <td>5.471267</td>\n",
       "      <td>4.992964</td>\n",
       "      <td>-11.475415</td>\n",
       "      <td>-10.572828</td>\n",
       "      <td>0.902586</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>covered</th>\n",
       "      <td>62</td>\n",
       "      <td>75</td>\n",
       "      <td>137</td>\n",
       "      <td>3.431136</td>\n",
       "      <td>2.460249</td>\n",
       "      <td>-11.703221</td>\n",
       "      <td>-10.805002</td>\n",
       "      <td>0.898219</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>voiced</th>\n",
       "      <td>66</td>\n",
       "      <td>80</td>\n",
       "      <td>146</td>\n",
       "      <td>5.286488</td>\n",
       "      <td>4.814398</td>\n",
       "      <td>-11.611298</td>\n",
       "      <td>-10.716192</td>\n",
       "      <td>0.895106</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rose</th>\n",
       "      <td>66</td>\n",
       "      <td>81</td>\n",
       "      <td>147</td>\n",
       "      <td>1.720520</td>\n",
       "      <td>1.849442</td>\n",
       "      <td>-11.593596</td>\n",
       "      <td>-10.716192</td>\n",
       "      <td>0.877404</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>performs</th>\n",
       "      <td>68</td>\n",
       "      <td>84</td>\n",
       "      <td>152</td>\n",
       "      <td>3.833541</td>\n",
       "      <td>3.776689</td>\n",
       "      <td>-11.541757</td>\n",
       "      <td>-10.673757</td>\n",
       "      <td>0.868000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>married</th>\n",
       "      <td>1147</td>\n",
       "      <td>1425</td>\n",
       "      <td>2572</td>\n",
       "      <td>4.865403</td>\n",
       "      <td>4.766142</td>\n",
       "      <td>-7.473390</td>\n",
       "      <td>-6.617375</td>\n",
       "      <td>0.856015</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fell</th>\n",
       "      <td>130</td>\n",
       "      <td>162</td>\n",
       "      <td>292</td>\n",
       "      <td>4.789238</td>\n",
       "      <td>4.197916</td>\n",
       "      <td>-10.602420</td>\n",
       "      <td>-9.748859</td>\n",
       "      <td>0.853561</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>donated</th>\n",
       "      <td>63</td>\n",
       "      <td>81</td>\n",
       "      <td>144</td>\n",
       "      <td>4.113383</td>\n",
       "      <td>3.644513</td>\n",
       "      <td>-11.593596</td>\n",
       "      <td>-10.782282</td>\n",
       "      <td>0.811314</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>portrayed</th>\n",
       "      <td>104</td>\n",
       "      <td>134</td>\n",
       "      <td>238</td>\n",
       "      <td>4.760440</td>\n",
       "      <td>4.393549</td>\n",
       "      <td>-10.874332</td>\n",
       "      <td>-10.068036</td>\n",
       "      <td>0.806296</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>performed</th>\n",
       "      <td>1581</td>\n",
       "      <td>2038</td>\n",
       "      <td>3619</td>\n",
       "      <td>5.066221</td>\n",
       "      <td>4.683320</td>\n",
       "      <td>-6.957502</td>\n",
       "      <td>-6.154748</td>\n",
       "      <td>0.802754</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>supported</th>\n",
       "      <td>94</td>\n",
       "      <td>124</td>\n",
       "      <td>218</td>\n",
       "      <td>4.251069</td>\n",
       "      <td>3.824166</td>\n",
       "      <td>-10.985364</td>\n",
       "      <td>-10.212426</td>\n",
       "      <td>0.772938</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>announced</th>\n",
       "      <td>239</td>\n",
       "      <td>321</td>\n",
       "      <td>560</td>\n",
       "      <td>3.196927</td>\n",
       "      <td>2.733603</td>\n",
       "      <td>-9.620231</td>\n",
       "      <td>-8.875391</td>\n",
       "      <td>0.744840</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gave</th>\n",
       "      <td>440</td>\n",
       "      <td>598</td>\n",
       "      <td>1038</td>\n",
       "      <td>4.813826</td>\n",
       "      <td>4.284739</td>\n",
       "      <td>-8.724736</td>\n",
       "      <td>-7.997647</td>\n",
       "      <td>0.727089</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>landed</th>\n",
       "      <td>77</td>\n",
       "      <td>107</td>\n",
       "      <td>184</td>\n",
       "      <td>5.281875</td>\n",
       "      <td>4.148207</td>\n",
       "      <td>-11.196261</td>\n",
       "      <td>-10.496879</td>\n",
       "      <td>0.699381</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>represented</th>\n",
       "      <td>103</td>\n",
       "      <td>145</td>\n",
       "      <td>248</td>\n",
       "      <td>3.660716</td>\n",
       "      <td>3.286384</td>\n",
       "      <td>-10.761324</td>\n",
       "      <td>-10.081842</td>\n",
       "      <td>0.679482</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>writes</th>\n",
       "      <td>77</td>\n",
       "      <td>109</td>\n",
       "      <td>186</td>\n",
       "      <td>4.508881</td>\n",
       "      <td>3.919714</td>\n",
       "      <td>-11.169788</td>\n",
       "      <td>-10.496879</td>\n",
       "      <td>0.672909</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>commented</th>\n",
       "      <td>53</td>\n",
       "      <td>76</td>\n",
       "      <td>129</td>\n",
       "      <td>3.672882</td>\n",
       "      <td>3.572109</td>\n",
       "      <td>-11.684362</td>\n",
       "      <td>-11.027394</td>\n",
       "      <td>0.656967</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>debuted</th>\n",
       "      <td>155</td>\n",
       "      <td>223</td>\n",
       "      <td>378</td>\n",
       "      <td>3.778022</td>\n",
       "      <td>3.882940</td>\n",
       "      <td>-10.143793</td>\n",
       "      <td>-9.496879</td>\n",
       "      <td>0.646914</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              she+word count  he+word count  total count  PMI she+word  \\\n",
       "placed                   836            155          991      5.852508   \n",
       "qualified                119             61          180      5.265387   \n",
       "sang                    1025            830         1855      5.552048   \n",
       "auditioned                90             74          164      5.452698   \n",
       "sings                    128            116          244      3.818992   \n",
       "starred                  438            412          850      5.037241   \n",
       "competed                 450            426          876      5.256862   \n",
       "photographed              81             85          166      4.371617   \n",
       "scored                   181            191          372      5.565560   \n",
       "finished                 391            416          807      5.481591   \n",
       "revealed                  93            101          194      3.366620   \n",
       "acted                     88            100          188      4.940183   \n",
       "wore                      73             88          161      5.471267   \n",
       "covered                   62             75          137      3.431136   \n",
       "voiced                    66             80          146      5.286488   \n",
       "rose                      66             81          147      1.720520   \n",
       "performs                  68             84          152      3.833541   \n",
       "married                 1147           1425         2572      4.865403   \n",
       "fell                     130            162          292      4.789238   \n",
       "donated                   63             81          144      4.113383   \n",
       "portrayed                104            134          238      4.760440   \n",
       "performed               1581           2038         3619      5.066221   \n",
       "supported                 94            124          218      4.251069   \n",
       "announced                239            321          560      3.196927   \n",
       "gave                     440            598         1038      4.813826   \n",
       "landed                    77            107          184      5.281875   \n",
       "represented              103            145          248      3.660716   \n",
       "writes                    77            109          186      4.508881   \n",
       "commented                 53             76          129      3.672882   \n",
       "debuted                  155            223          378      3.778022   \n",
       "\n",
       "              PMI he+word  log prob of word | he  log prob of word | she  \\\n",
       "placed           3.584022             -10.665746               -7.073198   \n",
       "qualified        4.799512             -11.996952               -9.875391   \n",
       "sang             5.252919              -8.252443               -6.779467   \n",
       "auditioned       5.192511             -11.722329              -10.274487   \n",
       "sings            3.368039             -11.080783               -9.771054   \n",
       "starred          4.752705              -9.261150               -8.004204   \n",
       "competed         4.774455              -9.213056               -7.965298   \n",
       "photographed     4.309556             -11.524883              -10.424730   \n",
       "scored           4.683200             -10.366186               -9.274487   \n",
       "finished         4.766644              -9.247245               -8.167572   \n",
       "revealed         2.599228             -11.278723              -10.227693   \n",
       "acted            4.563683             -11.292937              -10.306548   \n",
       "wore             4.992964             -11.475415              -10.572828   \n",
       "covered          2.460249             -11.703221              -10.805002   \n",
       "voiced           4.814398             -11.611298              -10.716192   \n",
       "rose             1.849442             -11.593596              -10.716192   \n",
       "performs         3.776689             -11.541757              -10.673757   \n",
       "married          4.766142              -7.473390               -6.617375   \n",
       "fell             4.197916             -10.602420               -9.748859   \n",
       "donated          3.644513             -11.593596              -10.782282   \n",
       "portrayed        4.393549             -10.874332              -10.068036   \n",
       "performed        4.683320              -6.957502               -6.154748   \n",
       "supported        3.824166             -10.985364              -10.212426   \n",
       "announced        2.733603              -9.620231               -8.875391   \n",
       "gave             4.284739              -8.724736               -7.997647   \n",
       "landed           4.148207             -11.196261              -10.496879   \n",
       "represented      3.286384             -10.761324              -10.081842   \n",
       "writes           3.919714             -11.169788              -10.496879   \n",
       "commented        3.572109             -11.684362              -11.027394   \n",
       "debuted          3.882940             -10.143793               -9.496879   \n",
       "\n",
       "                   LLR  \n",
       "placed        3.592548  \n",
       "qualified     2.121561  \n",
       "sang          1.472977  \n",
       "auditioned    1.447842  \n",
       "sings         1.309729  \n",
       "starred       1.256946  \n",
       "competed      1.247758  \n",
       "photographed  1.100154  \n",
       "scored        1.091699  \n",
       "finished      1.079673  \n",
       "revealed      1.051030  \n",
       "acted         0.986388  \n",
       "wore          0.902586  \n",
       "covered       0.898219  \n",
       "voiced        0.895106  \n",
       "rose          0.877404  \n",
       "performs      0.868000  \n",
       "married       0.856015  \n",
       "fell          0.853561  \n",
       "donated       0.811314  \n",
       "portrayed     0.806296  \n",
       "performed     0.802754  \n",
       "supported     0.772938  \n",
       "announced     0.744840  \n",
       "gave          0.727089  \n",
       "landed        0.699381  \n",
       "represented   0.679482  \n",
       "writes        0.672909  \n",
       "commented     0.656967  \n",
       "debuted       0.646914  "
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "she_words.head(30)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 972x936 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "she_top = she_words.drop(she_words[she_words['she+word count'] < 75 ].index).head(15)['LLR']\n",
    "Y1, X1 = she_top.index, she_top \n",
    "Y2, X2 = he_words.tail(15).index, he_words.tail(15)['LLR']\n",
    "\n",
    "title=\"Log likelihood ratio that word appears after she/he in articles about female and male artists, respectively\"\n",
    "subtitle=\"Limited to words appearing at least 75 times after pronoun and with positive word association (PMI) for that gender\"\n",
    "subtitle2=\"\\n\\nHe...                                       She...\"\n",
    "\n",
    "Y = np.arange(30)\n",
    "X_women = np.concatenate((np.zeros(15), np.flip(X1.values, 0)), axis=0)\n",
    "X_men = np.concatenate((np.flip(X2.values, 0), np.zeros(15)), axis=0)\n",
    "\n",
    "plt.figure(figsize=(13.5,13))\n",
    "plt.axes([0.025, 0.025, 0.95, 0.95])\n",
    "plt.barh(Y, X_women, facecolor=\"None\", edgecolor='#2ca79b')\n",
    "plt.barh(Y, X_men, facecolor=\"None\", edgecolor='#a52e3b')\n",
    "\n",
    "for x, y, label in zip(X_women, Y, np.concatenate((np.empty(15, dtype=\"U\") , np.flip(Y1,0)))):\n",
    "    plt.text(x+2.8, y, label, ha='right', va= 'center', color=\"black\", fontsize=13)\n",
    "for x, y, label in zip(X_men, Y, np.concatenate((np.flip(Y2,0), np.empty(15, dtype=\"U\")))):\n",
    "    plt.text(x-2, y, label, ha='left', va= 'center', color=\"black\", fontsize=13)\n",
    "\n",
    "\n",
    "plt.title(title, fontsize=17, y=1.06)\n",
    "plt.suptitle(subtitle + subtitle2, fontsize=14, y=1.03)\n",
    "plt.axvline(x=0, color=\"black\")\n",
    "plt.ylim(-.5, 29.6)\n",
    "plt.yticks(())\n",
    "plt.xlim(-12, 13.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Category: Sports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "scrolled": true
   },
   "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>she+word count</th>\n",
       "      <th>he+word count</th>\n",
       "      <th>total count</th>\n",
       "      <th>PMI she+word</th>\n",
       "      <th>PMI he+word</th>\n",
       "      <th>log prob of word | he</th>\n",
       "      <th>log prob of word | she</th>\n",
       "      <th>LLR</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>coached</th>\n",
       "      <td>29</td>\n",
       "      <td>350</td>\n",
       "      <td>379</td>\n",
       "      <td>2.750633</td>\n",
       "      <td>4.482452</td>\n",
       "      <td>-9.036847</td>\n",
       "      <td>-10.528389</td>\n",
       "      <td>-1.491542</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>resigned</th>\n",
       "      <td>10</td>\n",
       "      <td>128</td>\n",
       "      <td>138</td>\n",
       "      <td>5.076819</td>\n",
       "      <td>5.059008</td>\n",
       "      <td>-10.480947</td>\n",
       "      <td>-11.975848</td>\n",
       "      <td>-1.494901</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rejoined</th>\n",
       "      <td>7</td>\n",
       "      <td>94</td>\n",
       "      <td>101</td>\n",
       "      <td>5.521604</td>\n",
       "      <td>5.525421</td>\n",
       "      <td>-10.922318</td>\n",
       "      <td>-12.435280</td>\n",
       "      <td>-1.512961</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>might</th>\n",
       "      <td>7</td>\n",
       "      <td>97</td>\n",
       "      <td>104</td>\n",
       "      <td>4.265265</td>\n",
       "      <td>4.920743</td>\n",
       "      <td>-10.877464</td>\n",
       "      <td>-12.435280</td>\n",
       "      <td>-1.557815</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>stayed</th>\n",
       "      <td>20</td>\n",
       "      <td>258</td>\n",
       "      <td>278</td>\n",
       "      <td>5.348121</td>\n",
       "      <td>5.654440</td>\n",
       "      <td>-9.475366</td>\n",
       "      <td>-11.042962</td>\n",
       "      <td>-1.567596</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>refused</th>\n",
       "      <td>7</td>\n",
       "      <td>98</td>\n",
       "      <td>105</td>\n",
       "      <td>3.462711</td>\n",
       "      <td>4.180518</td>\n",
       "      <td>-10.862817</td>\n",
       "      <td>-12.435280</td>\n",
       "      <td>-1.572462</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>remained</th>\n",
       "      <td>40</td>\n",
       "      <td>508</td>\n",
       "      <td>548</td>\n",
       "      <td>4.958175</td>\n",
       "      <td>5.343527</td>\n",
       "      <td>-8.500652</td>\n",
       "      <td>-10.077728</td>\n",
       "      <td>-1.577075</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>picked</th>\n",
       "      <td>10</td>\n",
       "      <td>139</td>\n",
       "      <td>149</td>\n",
       "      <td>4.097578</td>\n",
       "      <td>4.284864</td>\n",
       "      <td>-10.362891</td>\n",
       "      <td>-11.975848</td>\n",
       "      <td>-1.612957</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>featured</th>\n",
       "      <td>12</td>\n",
       "      <td>167</td>\n",
       "      <td>179</td>\n",
       "      <td>2.142707</td>\n",
       "      <td>3.874006</td>\n",
       "      <td>-10.099857</td>\n",
       "      <td>-11.734840</td>\n",
       "      <td>-1.634983</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>impressed</th>\n",
       "      <td>5</td>\n",
       "      <td>77</td>\n",
       "      <td>82</td>\n",
       "      <td>4.451215</td>\n",
       "      <td>4.394064</td>\n",
       "      <td>-11.206772</td>\n",
       "      <td>-12.850317</td>\n",
       "      <td>-1.643545</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>called</th>\n",
       "      <td>7</td>\n",
       "      <td>104</td>\n",
       "      <td>111</td>\n",
       "      <td>1.236202</td>\n",
       "      <td>1.853365</td>\n",
       "      <td>-10.777928</td>\n",
       "      <td>-12.435280</td>\n",
       "      <td>-1.657351</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>saw</th>\n",
       "      <td>11</td>\n",
       "      <td>162</td>\n",
       "      <td>173</td>\n",
       "      <td>2.277517</td>\n",
       "      <td>3.082396</td>\n",
       "      <td>-10.143446</td>\n",
       "      <td>-11.850317</td>\n",
       "      <td>-1.706871</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>left</th>\n",
       "      <td>78</td>\n",
       "      <td>1075</td>\n",
       "      <td>1153</td>\n",
       "      <td>0.474783</td>\n",
       "      <td>1.930364</td>\n",
       "      <td>-7.420712</td>\n",
       "      <td>-9.131499</td>\n",
       "      <td>-1.710787</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>joined</th>\n",
       "      <td>169</td>\n",
       "      <td>2334</td>\n",
       "      <td>2503</td>\n",
       "      <td>5.065885</td>\n",
       "      <td>5.416196</td>\n",
       "      <td>-6.302967</td>\n",
       "      <td>-8.025889</td>\n",
       "      <td>-1.722922</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>replaced</th>\n",
       "      <td>10</td>\n",
       "      <td>152</td>\n",
       "      <td>162</td>\n",
       "      <td>3.519602</td>\n",
       "      <td>3.864131</td>\n",
       "      <td>-10.234786</td>\n",
       "      <td>-11.975848</td>\n",
       "      <td>-1.741062</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>can</th>\n",
       "      <td>22</td>\n",
       "      <td>326</td>\n",
       "      <td>348</td>\n",
       "      <td>1.408441</td>\n",
       "      <td>3.192723</td>\n",
       "      <td>-9.139027</td>\n",
       "      <td>-10.911718</td>\n",
       "      <td>-1.772690</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>scored</th>\n",
       "      <td>376</td>\n",
       "      <td>5762</td>\n",
       "      <td>6138</td>\n",
       "      <td>5.488936</td>\n",
       "      <td>5.598241</td>\n",
       "      <td>-4.999570</td>\n",
       "      <td>-6.876859</td>\n",
       "      <td>-1.877289</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>spent</th>\n",
       "      <td>66</td>\n",
       "      <td>1048</td>\n",
       "      <td>1114</td>\n",
       "      <td>5.242068</td>\n",
       "      <td>5.489896</td>\n",
       "      <td>-7.457375</td>\n",
       "      <td>-9.369190</td>\n",
       "      <td>-1.911815</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>drove</th>\n",
       "      <td>4</td>\n",
       "      <td>78</td>\n",
       "      <td>82</td>\n",
       "      <td>4.360612</td>\n",
       "      <td>4.825600</td>\n",
       "      <td>-11.188393</td>\n",
       "      <td>-13.113351</td>\n",
       "      <td>-1.924958</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>re</th>\n",
       "      <td>6</td>\n",
       "      <td>115</td>\n",
       "      <td>121</td>\n",
       "      <td>1.526622</td>\n",
       "      <td>2.517965</td>\n",
       "      <td>-10.634193</td>\n",
       "      <td>-12.627925</td>\n",
       "      <td>-1.993732</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>amassed</th>\n",
       "      <td>4</td>\n",
       "      <td>85</td>\n",
       "      <td>89</td>\n",
       "      <td>4.884174</td>\n",
       "      <td>5.912005</td>\n",
       "      <td>-11.065909</td>\n",
       "      <td>-13.113351</td>\n",
       "      <td>-2.047442</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>believed</th>\n",
       "      <td>5</td>\n",
       "      <td>123</td>\n",
       "      <td>128</td>\n",
       "      <td>4.118640</td>\n",
       "      <td>4.696668</td>\n",
       "      <td>-10.537978</td>\n",
       "      <td>-12.850317</td>\n",
       "      <td>-2.312339</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>formed</th>\n",
       "      <td>3</td>\n",
       "      <td>98</td>\n",
       "      <td>101</td>\n",
       "      <td>2.945575</td>\n",
       "      <td>4.441831</td>\n",
       "      <td>-10.862817</td>\n",
       "      <td>-13.435280</td>\n",
       "      <td>-2.572462</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>signed</th>\n",
       "      <td>102</td>\n",
       "      <td>2727</td>\n",
       "      <td>2829</td>\n",
       "      <td>4.579320</td>\n",
       "      <td>4.723181</td>\n",
       "      <td>-6.078546</td>\n",
       "      <td>-8.748779</td>\n",
       "      <td>-2.670233</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>compiled</th>\n",
       "      <td>3</td>\n",
       "      <td>109</td>\n",
       "      <td>112</td>\n",
       "      <td>3.611156</td>\n",
       "      <td>5.222164</td>\n",
       "      <td>-10.710814</td>\n",
       "      <td>-13.435280</td>\n",
       "      <td>-2.724465</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>hit</th>\n",
       "      <td>18</td>\n",
       "      <td>558</td>\n",
       "      <td>576</td>\n",
       "      <td>3.076819</td>\n",
       "      <td>4.336480</td>\n",
       "      <td>-8.365469</td>\n",
       "      <td>-11.187352</td>\n",
       "      <td>-2.821883</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>caught</th>\n",
       "      <td>5</td>\n",
       "      <td>180</td>\n",
       "      <td>185</td>\n",
       "      <td>3.228823</td>\n",
       "      <td>4.465539</td>\n",
       "      <td>-9.992328</td>\n",
       "      <td>-12.850317</td>\n",
       "      <td>-2.857989</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>netted</th>\n",
       "      <td>4</td>\n",
       "      <td>179</td>\n",
       "      <td>183</td>\n",
       "      <td>4.884174</td>\n",
       "      <td>5.925889</td>\n",
       "      <td>-10.000321</td>\n",
       "      <td>-13.113351</td>\n",
       "      <td>-3.113031</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>batted</th>\n",
       "      <td>7</td>\n",
       "      <td>327</td>\n",
       "      <td>334</td>\n",
       "      <td>4.265265</td>\n",
       "      <td>5.159768</td>\n",
       "      <td>-9.134622</td>\n",
       "      <td>-12.435280</td>\n",
       "      <td>-3.300658</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pitched</th>\n",
       "      <td>0</td>\n",
       "      <td>152</td>\n",
       "      <td>152</td>\n",
       "      <td>-100.000000</td>\n",
       "      <td>4.930290</td>\n",
       "      <td>-10.234786</td>\n",
       "      <td>-15.435280</td>\n",
       "      <td>-5.200493</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           she+word count  he+word count  total count  PMI she+word  \\\n",
       "coached                29            350          379      2.750633   \n",
       "resigned               10            128          138      5.076819   \n",
       "rejoined                7             94          101      5.521604   \n",
       "might                   7             97          104      4.265265   \n",
       "stayed                 20            258          278      5.348121   \n",
       "refused                 7             98          105      3.462711   \n",
       "remained               40            508          548      4.958175   \n",
       "picked                 10            139          149      4.097578   \n",
       "featured               12            167          179      2.142707   \n",
       "impressed               5             77           82      4.451215   \n",
       "called                  7            104          111      1.236202   \n",
       "saw                    11            162          173      2.277517   \n",
       "left                   78           1075         1153      0.474783   \n",
       "joined                169           2334         2503      5.065885   \n",
       "replaced               10            152          162      3.519602   \n",
       "can                    22            326          348      1.408441   \n",
       "scored                376           5762         6138      5.488936   \n",
       "spent                  66           1048         1114      5.242068   \n",
       "drove                   4             78           82      4.360612   \n",
       "re                      6            115          121      1.526622   \n",
       "amassed                 4             85           89      4.884174   \n",
       "believed                5            123          128      4.118640   \n",
       "formed                  3             98          101      2.945575   \n",
       "signed                102           2727         2829      4.579320   \n",
       "compiled                3            109          112      3.611156   \n",
       "hit                    18            558          576      3.076819   \n",
       "caught                  5            180          185      3.228823   \n",
       "netted                  4            179          183      4.884174   \n",
       "batted                  7            327          334      4.265265   \n",
       "pitched                 0            152          152   -100.000000   \n",
       "\n",
       "           PMI he+word  log prob of word | he  log prob of word | she  \\\n",
       "coached       4.482452              -9.036847              -10.528389   \n",
       "resigned      5.059008             -10.480947              -11.975848   \n",
       "rejoined      5.525421             -10.922318              -12.435280   \n",
       "might         4.920743             -10.877464              -12.435280   \n",
       "stayed        5.654440              -9.475366              -11.042962   \n",
       "refused       4.180518             -10.862817              -12.435280   \n",
       "remained      5.343527              -8.500652              -10.077728   \n",
       "picked        4.284864             -10.362891              -11.975848   \n",
       "featured      3.874006             -10.099857              -11.734840   \n",
       "impressed     4.394064             -11.206772              -12.850317   \n",
       "called        1.853365             -10.777928              -12.435280   \n",
       "saw           3.082396             -10.143446              -11.850317   \n",
       "left          1.930364              -7.420712               -9.131499   \n",
       "joined        5.416196              -6.302967               -8.025889   \n",
       "replaced      3.864131             -10.234786              -11.975848   \n",
       "can           3.192723              -9.139027              -10.911718   \n",
       "scored        5.598241              -4.999570               -6.876859   \n",
       "spent         5.489896              -7.457375               -9.369190   \n",
       "drove         4.825600             -11.188393              -13.113351   \n",
       "re            2.517965             -10.634193              -12.627925   \n",
       "amassed       5.912005             -11.065909              -13.113351   \n",
       "believed      4.696668             -10.537978              -12.850317   \n",
       "formed        4.441831             -10.862817              -13.435280   \n",
       "signed        4.723181              -6.078546               -8.748779   \n",
       "compiled      5.222164             -10.710814              -13.435280   \n",
       "hit           4.336480              -8.365469              -11.187352   \n",
       "caught        4.465539              -9.992328              -12.850317   \n",
       "netted        5.925889             -10.000321              -13.113351   \n",
       "batted        5.159768              -9.134622              -12.435280   \n",
       "pitched       4.930290             -10.234786              -15.435280   \n",
       "\n",
       "                LLR  \n",
       "coached   -1.491542  \n",
       "resigned  -1.494901  \n",
       "rejoined  -1.512961  \n",
       "might     -1.557815  \n",
       "stayed    -1.567596  \n",
       "refused   -1.572462  \n",
       "remained  -1.577075  \n",
       "picked    -1.612957  \n",
       "featured  -1.634983  \n",
       "impressed -1.643545  \n",
       "called    -1.657351  \n",
       "saw       -1.706871  \n",
       "left      -1.710787  \n",
       "joined    -1.722922  \n",
       "replaced  -1.741062  \n",
       "can       -1.772690  \n",
       "scored    -1.877289  \n",
       "spent     -1.911815  \n",
       "drove     -1.924958  \n",
       "re        -1.993732  \n",
       "amassed   -2.047442  \n",
       "believed  -2.312339  \n",
       "formed    -2.572462  \n",
       "signed    -2.670233  \n",
       "compiled  -2.724465  \n",
       "hit       -2.821883  \n",
       "caught    -2.857989  \n",
       "netted    -3.113031  \n",
       "batted    -3.300658  \n",
       "pitched   -5.200493  "
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Number of words in corpus of men and women's SPORTS bios: \n",
    "n_women = sum(female_sports_u.values())\n",
    "n_men = sum(male_sports_u.values())\n",
    "\n",
    "# bigrams beginning with he or she, filtered by words appearing at least 75 times.\n",
    "she = collections.Counter({ k[1]: v for k, v in female_sports.items() if v >= 75 and k[0] == \"she\" })\n",
    "he = collections.Counter({ k[1]: v for k, v in male_sports.items() if v >= 75 and k[0] == \"he\" })\n",
    "\n",
    "he_count = male_sports_u['he']\n",
    "she_count = female_sports_u['she']\n",
    "\n",
    "for word in he:\n",
    "    if word not in she:\n",
    "        she[word] = female_sports[(\"she\", word)]\n",
    "for word in she:\n",
    "    if word not in he:\n",
    "        he[word] = male_sports[(\"he\", word)]\n",
    "\n",
    "columns = ['Word following he/she', \"she+word count\"]\n",
    "df = pd.DataFrame([(k,v) for k,v in she.items()], columns=columns)\n",
    "df.set_index(['Word following he/she'], inplace=True)\n",
    "columns2 = ['Word following he/she', \"he+word count\",]\n",
    "df2 = pd.DataFrame([(k,v) for k,v in he.items()], columns=columns2)\n",
    "df2.set_index(['Word following he/she'], inplace=True)\n",
    "\n",
    "df = pd.concat([df,df2], axis=1)\n",
    "df['total count'] = df.fillna(0)[\"she+word count\"] + df.fillna(0)[\"he+word count\"]\n",
    "df.sort_values(by=['total count'], ascending=False, inplace=True)\n",
    "\n",
    "df['PMI she+word'] = df.apply(lambda row: pmi(she_count, female_sports_u[row.name], \n",
    "                                              row[\"she+word count\"], n_women), axis=1)\n",
    "df['PMI he+word'] = df.apply(lambda row: pmi(he_count, male_sports_u[row.name], \n",
    "                                             row[\"he+word count\"], n_men), axis=1)\n",
    "\n",
    "# Log probabilities of word given he/she. Add 1 for smoothing\n",
    "df['log prob of word | he'] = np.log2(\n",
    "    (df.fillna(0)[\"he+word count\"] + 1)) - np.log2((sum(he.values()) + 1))\n",
    "df['log prob of word | she'] = np.log2(\n",
    "    (df.fillna(0)[\"she+word count\"] + 1)) - np.log2((sum(she.values()) + 1))\n",
    "\n",
    "# Log likelihood ratio she : he\n",
    "df['LLR'] = df['log prob of word | she'] - df['log prob of word | he']\n",
    "df.sort_values(by=['LLR'], ascending=False, inplace=True)\n",
    "\n",
    "he_words = df.drop(df[df['PMI he+word'] <= 0 ].index).tail(100)\n",
    "she_words = df.drop(df[df['PMI she+word'] <= 0 ].index).head(100)\n",
    "he_words.tail(30)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "he_words.drop(['re'], axis=0, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "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>she+word count</th>\n",
       "      <th>he+word count</th>\n",
       "      <th>total count</th>\n",
       "      <th>PMI she+word</th>\n",
       "      <th>PMI he+word</th>\n",
       "      <th>log prob of word | he</th>\n",
       "      <th>log prob of word | she</th>\n",
       "      <th>LLR</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>placed</th>\n",
       "      <td>699</td>\n",
       "      <td>769</td>\n",
       "      <td>1468</td>\n",
       "      <td>5.688565</td>\n",
       "      <td>5.021521</td>\n",
       "      <td>-7.903459</td>\n",
       "      <td>-5.984068</td>\n",
       "      <td>1.919391</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>reached</th>\n",
       "      <td>512</td>\n",
       "      <td>675</td>\n",
       "      <td>1187</td>\n",
       "      <td>5.645770</td>\n",
       "      <td>4.913777</td>\n",
       "      <td>-8.091295</td>\n",
       "      <td>-6.432465</td>\n",
       "      <td>1.658830</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>competes</th>\n",
       "      <td>59</td>\n",
       "      <td>82</td>\n",
       "      <td>141</td>\n",
       "      <td>4.094392</td>\n",
       "      <td>4.363010</td>\n",
       "      <td>-11.117135</td>\n",
       "      <td>-9.528389</td>\n",
       "      <td>1.588746</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>married</th>\n",
       "      <td>222</td>\n",
       "      <td>311</td>\n",
       "      <td>533</td>\n",
       "      <td>4.803609</td>\n",
       "      <td>4.104026</td>\n",
       "      <td>-9.206772</td>\n",
       "      <td>-7.634380</td>\n",
       "      <td>1.572392</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>improved</th>\n",
       "      <td>201</td>\n",
       "      <td>285</td>\n",
       "      <td>486</td>\n",
       "      <td>5.555086</td>\n",
       "      <td>4.957627</td>\n",
       "      <td>-9.332303</td>\n",
       "      <td>-7.777068</td>\n",
       "      <td>1.555235</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>jumped</th>\n",
       "      <td>67</td>\n",
       "      <td>97</td>\n",
       "      <td>164</td>\n",
       "      <td>5.873448</td>\n",
       "      <td>5.086133</td>\n",
       "      <td>-10.877464</td>\n",
       "      <td>-9.347817</td>\n",
       "      <td>1.529647</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>qualified</th>\n",
       "      <td>249</td>\n",
       "      <td>378</td>\n",
       "      <td>627</td>\n",
       "      <td>5.193125</td>\n",
       "      <td>4.798392</td>\n",
       "      <td>-8.926120</td>\n",
       "      <td>-7.469495</td>\n",
       "      <td>1.456625</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ran</th>\n",
       "      <td>753</td>\n",
       "      <td>1226</td>\n",
       "      <td>1979</td>\n",
       "      <td>6.044076</td>\n",
       "      <td>5.506641</td>\n",
       "      <td>-7.231254</td>\n",
       "      <td>-5.876859</td>\n",
       "      <td>1.354396</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ranked</th>\n",
       "      <td>57</td>\n",
       "      <td>98</td>\n",
       "      <td>155</td>\n",
       "      <td>2.984049</td>\n",
       "      <td>2.379947</td>\n",
       "      <td>-10.862817</td>\n",
       "      <td>-9.577299</td>\n",
       "      <td>1.285519</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>lives</th>\n",
       "      <td>64</td>\n",
       "      <td>110</td>\n",
       "      <td>174</td>\n",
       "      <td>4.689417</td>\n",
       "      <td>4.157757</td>\n",
       "      <td>-10.697758</td>\n",
       "      <td>-9.412912</td>\n",
       "      <td>1.284846</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>beat</th>\n",
       "      <td>214</td>\n",
       "      <td>368</td>\n",
       "      <td>582</td>\n",
       "      <td>5.016463</td>\n",
       "      <td>4.105611</td>\n",
       "      <td>-8.964697</td>\n",
       "      <td>-7.687087</td>\n",
       "      <td>1.277610</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>defeated</th>\n",
       "      <td>366</td>\n",
       "      <td>632</td>\n",
       "      <td>998</td>\n",
       "      <td>5.058911</td>\n",
       "      <td>4.312894</td>\n",
       "      <td>-8.186112</td>\n",
       "      <td>-6.915643</td>\n",
       "      <td>1.270469</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>represented</th>\n",
       "      <td>556</td>\n",
       "      <td>1009</td>\n",
       "      <td>1565</td>\n",
       "      <td>5.462019</td>\n",
       "      <td>5.167664</td>\n",
       "      <td>-7.512034</td>\n",
       "      <td>-6.313746</td>\n",
       "      <td>1.198288</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>set</th>\n",
       "      <td>591</td>\n",
       "      <td>1173</td>\n",
       "      <td>1764</td>\n",
       "      <td>4.772646</td>\n",
       "      <td>4.328836</td>\n",
       "      <td>-7.294957</td>\n",
       "      <td>-6.225826</td>\n",
       "      <td>1.069131</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>won</th>\n",
       "      <td>4766</td>\n",
       "      <td>9479</td>\n",
       "      <td>14245</td>\n",
       "      <td>5.644844</td>\n",
       "      <td>5.284666</td>\n",
       "      <td>-4.281503</td>\n",
       "      <td>-3.216414</td>\n",
       "      <td>1.065089</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>holds</th>\n",
       "      <td>145</td>\n",
       "      <td>293</td>\n",
       "      <td>438</td>\n",
       "      <td>5.346407</td>\n",
       "      <td>4.765958</td>\n",
       "      <td>-9.292502</td>\n",
       "      <td>-8.245455</td>\n",
       "      <td>1.047047</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>entered</th>\n",
       "      <td>182</td>\n",
       "      <td>370</td>\n",
       "      <td>552</td>\n",
       "      <td>5.999652</td>\n",
       "      <td>5.116701</td>\n",
       "      <td>-8.956899</td>\n",
       "      <td>-7.919580</td>\n",
       "      <td>1.037319</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fell</th>\n",
       "      <td>83</td>\n",
       "      <td>170</td>\n",
       "      <td>253</td>\n",
       "      <td>5.001826</td>\n",
       "      <td>4.544634</td>\n",
       "      <td>-10.074321</td>\n",
       "      <td>-9.042962</td>\n",
       "      <td>1.031359</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>previously</th>\n",
       "      <td>77</td>\n",
       "      <td>158</td>\n",
       "      <td>235</td>\n",
       "      <td>5.037219</td>\n",
       "      <td>3.825064</td>\n",
       "      <td>-10.179291</td>\n",
       "      <td>-9.149877</td>\n",
       "      <td>1.029414</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cleared</th>\n",
       "      <td>46</td>\n",
       "      <td>95</td>\n",
       "      <td>141</td>\n",
       "      <td>4.940131</td>\n",
       "      <td>4.076598</td>\n",
       "      <td>-10.907211</td>\n",
       "      <td>-9.880691</td>\n",
       "      <td>1.026521</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>competed</th>\n",
       "      <td>1750</td>\n",
       "      <td>3641</td>\n",
       "      <td>5391</td>\n",
       "      <td>5.646086</td>\n",
       "      <td>5.666464</td>\n",
       "      <td>-5.661659</td>\n",
       "      <td>-4.661316</td>\n",
       "      <td>1.000343</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>lost</th>\n",
       "      <td>656</td>\n",
       "      <td>1426</td>\n",
       "      <td>2082</td>\n",
       "      <td>5.291170</td>\n",
       "      <td>4.604638</td>\n",
       "      <td>-7.013404</td>\n",
       "      <td>-6.075530</td>\n",
       "      <td>0.937874</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gave</th>\n",
       "      <td>77</td>\n",
       "      <td>169</td>\n",
       "      <td>246</td>\n",
       "      <td>4.750081</td>\n",
       "      <td>3.669956</td>\n",
       "      <td>-10.082783</td>\n",
       "      <td>-9.149877</td>\n",
       "      <td>0.932906</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>trained</th>\n",
       "      <td>107</td>\n",
       "      <td>235</td>\n",
       "      <td>342</td>\n",
       "      <td>5.162117</td>\n",
       "      <td>4.980156</td>\n",
       "      <td>-9.609531</td>\n",
       "      <td>-8.680392</td>\n",
       "      <td>0.929139</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>finished</th>\n",
       "      <td>1873</td>\n",
       "      <td>4368</td>\n",
       "      <td>6241</td>\n",
       "      <td>5.852181</td>\n",
       "      <td>5.533595</td>\n",
       "      <td>-5.399087</td>\n",
       "      <td>-4.563374</td>\n",
       "      <td>0.835712</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>claimed</th>\n",
       "      <td>105</td>\n",
       "      <td>248</td>\n",
       "      <td>353</td>\n",
       "      <td>5.489895</td>\n",
       "      <td>4.547175</td>\n",
       "      <td>-9.532172</td>\n",
       "      <td>-8.707359</td>\n",
       "      <td>0.824813</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>performed</th>\n",
       "      <td>63</td>\n",
       "      <td>150</td>\n",
       "      <td>213</td>\n",
       "      <td>4.895670</td>\n",
       "      <td>4.842794</td>\n",
       "      <td>-10.253769</td>\n",
       "      <td>-9.435280</td>\n",
       "      <td>0.818490</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>achieved</th>\n",
       "      <td>139</td>\n",
       "      <td>331</td>\n",
       "      <td>470</td>\n",
       "      <td>3.867406</td>\n",
       "      <td>4.154588</td>\n",
       "      <td>-9.117135</td>\n",
       "      <td>-8.305997</td>\n",
       "      <td>0.811138</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>broke</th>\n",
       "      <td>233</td>\n",
       "      <td>587</td>\n",
       "      <td>820</td>\n",
       "      <td>5.465272</td>\n",
       "      <td>5.173762</td>\n",
       "      <td>-8.292502</td>\n",
       "      <td>-7.564915</td>\n",
       "      <td>0.727587</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>changed</th>\n",
       "      <td>36</td>\n",
       "      <td>93</td>\n",
       "      <td>129</td>\n",
       "      <td>4.815695</td>\n",
       "      <td>3.743115</td>\n",
       "      <td>-10.937585</td>\n",
       "      <td>-10.225826</td>\n",
       "      <td>0.711759</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             she+word count  he+word count  total count  PMI she+word  \\\n",
       "placed                  699            769         1468      5.688565   \n",
       "reached                 512            675         1187      5.645770   \n",
       "competes                 59             82          141      4.094392   \n",
       "married                 222            311          533      4.803609   \n",
       "improved                201            285          486      5.555086   \n",
       "jumped                   67             97          164      5.873448   \n",
       "qualified               249            378          627      5.193125   \n",
       "ran                     753           1226         1979      6.044076   \n",
       "ranked                   57             98          155      2.984049   \n",
       "lives                    64            110          174      4.689417   \n",
       "beat                    214            368          582      5.016463   \n",
       "defeated                366            632          998      5.058911   \n",
       "represented             556           1009         1565      5.462019   \n",
       "set                     591           1173         1764      4.772646   \n",
       "won                    4766           9479        14245      5.644844   \n",
       "holds                   145            293          438      5.346407   \n",
       "entered                 182            370          552      5.999652   \n",
       "fell                     83            170          253      5.001826   \n",
       "previously               77            158          235      5.037219   \n",
       "cleared                  46             95          141      4.940131   \n",
       "competed               1750           3641         5391      5.646086   \n",
       "lost                    656           1426         2082      5.291170   \n",
       "gave                     77            169          246      4.750081   \n",
       "trained                 107            235          342      5.162117   \n",
       "finished               1873           4368         6241      5.852181   \n",
       "claimed                 105            248          353      5.489895   \n",
       "performed                63            150          213      4.895670   \n",
       "achieved                139            331          470      3.867406   \n",
       "broke                   233            587          820      5.465272   \n",
       "changed                  36             93          129      4.815695   \n",
       "\n",
       "             PMI he+word  log prob of word | he  log prob of word | she  \\\n",
       "placed          5.021521              -7.903459               -5.984068   \n",
       "reached         4.913777              -8.091295               -6.432465   \n",
       "competes        4.363010             -11.117135               -9.528389   \n",
       "married         4.104026              -9.206772               -7.634380   \n",
       "improved        4.957627              -9.332303               -7.777068   \n",
       "jumped          5.086133             -10.877464               -9.347817   \n",
       "qualified       4.798392              -8.926120               -7.469495   \n",
       "ran             5.506641              -7.231254               -5.876859   \n",
       "ranked          2.379947             -10.862817               -9.577299   \n",
       "lives           4.157757             -10.697758               -9.412912   \n",
       "beat            4.105611              -8.964697               -7.687087   \n",
       "defeated        4.312894              -8.186112               -6.915643   \n",
       "represented     5.167664              -7.512034               -6.313746   \n",
       "set             4.328836              -7.294957               -6.225826   \n",
       "won             5.284666              -4.281503               -3.216414   \n",
       "holds           4.765958              -9.292502               -8.245455   \n",
       "entered         5.116701              -8.956899               -7.919580   \n",
       "fell            4.544634             -10.074321               -9.042962   \n",
       "previously      3.825064             -10.179291               -9.149877   \n",
       "cleared         4.076598             -10.907211               -9.880691   \n",
       "competed        5.666464              -5.661659               -4.661316   \n",
       "lost            4.604638              -7.013404               -6.075530   \n",
       "gave            3.669956             -10.082783               -9.149877   \n",
       "trained         4.980156              -9.609531               -8.680392   \n",
       "finished        5.533595              -5.399087               -4.563374   \n",
       "claimed         4.547175              -9.532172               -8.707359   \n",
       "performed       4.842794             -10.253769               -9.435280   \n",
       "achieved        4.154588              -9.117135               -8.305997   \n",
       "broke           5.173762              -8.292502               -7.564915   \n",
       "changed         3.743115             -10.937585              -10.225826   \n",
       "\n",
       "                  LLR  \n",
       "placed       1.919391  \n",
       "reached      1.658830  \n",
       "competes     1.588746  \n",
       "married      1.572392  \n",
       "improved     1.555235  \n",
       "jumped       1.529647  \n",
       "qualified    1.456625  \n",
       "ran          1.354396  \n",
       "ranked       1.285519  \n",
       "lives        1.284846  \n",
       "beat         1.277610  \n",
       "defeated     1.270469  \n",
       "represented  1.198288  \n",
       "set          1.069131  \n",
       "won          1.065089  \n",
       "holds        1.047047  \n",
       "entered      1.037319  \n",
       "fell         1.031359  \n",
       "previously   1.029414  \n",
       "cleared      1.026521  \n",
       "competed     1.000343  \n",
       "lost         0.937874  \n",
       "gave         0.932906  \n",
       "trained      0.929139  \n",
       "finished     0.835712  \n",
       "claimed      0.824813  \n",
       "performed    0.818490  \n",
       "achieved     0.811138  \n",
       "broke        0.727587  \n",
       "changed      0.711759  "
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "she_words.head(30)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 792x792 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "she_top = she_words.drop(she_words[she_words['she+word count'] < 75 ].index).head(15)['LLR']\n",
    "Y1, X1 = she_top.index, she_top \n",
    "Y2, X2 = he_words.tail(15).index, he_words.tail(15)['LLR']\n",
    "\n",
    "title=\"Log likelihood ratio that word appears after she/he in sports articles about female and male, respectively\"\n",
    "subtitle=\"Limited to words appearing at least 75 times after pronoun and with positive word association (PMI) for that gender\"\n",
    "subtitle2=\"\\n\\nHe...                                       She...\"\n",
    "\n",
    "Y = np.arange(30)\n",
    "X_women = np.concatenate((np.zeros(15), np.flip(X1.values, 0)), axis=0)\n",
    "X_men = np.concatenate((np.flip(X2.values, 0), np.zeros(15)), axis=0)\n",
    "\n",
    "plt.figure(figsize=(11,11))\n",
    "plt.axes([0.025, 0.025, 0.95, 0.95])\n",
    "plt.barh(Y, X_women, facecolor=\"None\", edgecolor='#2ca79b')\n",
    "plt.barh(Y, X_men, facecolor=\"None\", edgecolor='#a52e3b')\n",
    "\n",
    "for x, y, label in zip(X_women, Y, np.concatenate((np.empty(15, dtype=\"U\") , np.flip(Y1,0)))):\n",
    "    plt.text(x+2.8, y, label, ha='right', va= 'center', color=\"black\", fontsize=13)\n",
    "for x, y, label in zip(X_men, Y, np.concatenate((np.flip(Y2,0), np.empty(15, dtype=\"U\")))):\n",
    "    plt.text(x-2.1, y, label, ha='left', va= 'center', color=\"black\", fontsize=13)\n",
    "\n",
    "plt.title(title, fontsize=17, y=1.07)\n",
    "plt.suptitle(subtitle + subtitle2, fontsize=14, y=1.04)\n",
    "plt.axvline(x=0, color=\"black\")\n",
    "plt.ylim(-.5, 29.6)\n",
    "plt.yticks(())\n",
    "plt.xlim(-12, 13.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Category: Programming"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "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>she+word count</th>\n",
       "      <th>he+word count</th>\n",
       "      <th>total count</th>\n",
       "      <th>PMI she+word</th>\n",
       "      <th>PMI he+word</th>\n",
       "      <th>log prob of word | he</th>\n",
       "      <th>log prob of word | she</th>\n",
       "      <th>LLR</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>won</th>\n",
       "      <td>3</td>\n",
       "      <td>56</td>\n",
       "      <td>59</td>\n",
       "      <td>4.772765</td>\n",
       "      <td>4.553110</td>\n",
       "      <td>-7.028585</td>\n",
       "      <td>-7.047124</td>\n",
       "      <td>-0.018539</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>joined</th>\n",
       "      <td>8</td>\n",
       "      <td>129</td>\n",
       "      <td>137</td>\n",
       "      <td>4.825232</td>\n",
       "      <td>5.589872</td>\n",
       "      <td>-5.839107</td>\n",
       "      <td>-5.877199</td>\n",
       "      <td>-0.038092</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>started</th>\n",
       "      <td>7</td>\n",
       "      <td>117</td>\n",
       "      <td>124</td>\n",
       "      <td>5.047625</td>\n",
       "      <td>5.226618</td>\n",
       "      <td>-5.978832</td>\n",
       "      <td>-6.047124</td>\n",
       "      <td>-0.068292</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>continued</th>\n",
       "      <td>2</td>\n",
       "      <td>44</td>\n",
       "      <td>46</td>\n",
       "      <td>4.673229</td>\n",
       "      <td>5.030582</td>\n",
       "      <td>-7.369622</td>\n",
       "      <td>-7.462161</td>\n",
       "      <td>-0.092540</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>called</th>\n",
       "      <td>1</td>\n",
       "      <td>30</td>\n",
       "      <td>31</td>\n",
       "      <td>1.865874</td>\n",
       "      <td>2.824946</td>\n",
       "      <td>-7.907278</td>\n",
       "      <td>-8.047124</td>\n",
       "      <td>-0.139846</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>developed</th>\n",
       "      <td>5</td>\n",
       "      <td>93</td>\n",
       "      <td>98</td>\n",
       "      <td>4.857654</td>\n",
       "      <td>4.672099</td>\n",
       "      <td>-6.306886</td>\n",
       "      <td>-6.462161</td>\n",
       "      <td>-0.155276</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>has</th>\n",
       "      <td>37</td>\n",
       "      <td>598</td>\n",
       "      <td>635</td>\n",
       "      <td>5.262096</td>\n",
       "      <td>5.159993</td>\n",
       "      <td>-3.635062</td>\n",
       "      <td>-3.799196</td>\n",
       "      <td>-0.164134</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>helped</th>\n",
       "      <td>2</td>\n",
       "      <td>49</td>\n",
       "      <td>51</td>\n",
       "      <td>4.535726</td>\n",
       "      <td>5.050397</td>\n",
       "      <td>-7.217618</td>\n",
       "      <td>-7.462161</td>\n",
       "      <td>-0.244543</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>announced</th>\n",
       "      <td>1</td>\n",
       "      <td>33</td>\n",
       "      <td>34</td>\n",
       "      <td>2.825232</td>\n",
       "      <td>3.274311</td>\n",
       "      <td>-7.774012</td>\n",
       "      <td>-8.047124</td>\n",
       "      <td>-0.273112</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>served</th>\n",
       "      <td>4</td>\n",
       "      <td>85</td>\n",
       "      <td>89</td>\n",
       "      <td>4.602840</td>\n",
       "      <td>5.130055</td>\n",
       "      <td>-6.435210</td>\n",
       "      <td>-6.725196</td>\n",
       "      <td>-0.289986</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>released</th>\n",
       "      <td>1</td>\n",
       "      <td>34</td>\n",
       "      <td>35</td>\n",
       "      <td>2.673229</td>\n",
       "      <td>2.961052</td>\n",
       "      <td>-7.732192</td>\n",
       "      <td>-8.047124</td>\n",
       "      <td>-0.314932</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>is</th>\n",
       "      <td>54</td>\n",
       "      <td>974</td>\n",
       "      <td>1028</td>\n",
       "      <td>4.469274</td>\n",
       "      <td>4.585574</td>\n",
       "      <td>-2.932216</td>\n",
       "      <td>-3.265764</td>\n",
       "      <td>-0.333548</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>received</th>\n",
       "      <td>9</td>\n",
       "      <td>183</td>\n",
       "      <td>192</td>\n",
       "      <td>4.955629</td>\n",
       "      <td>5.504382</td>\n",
       "      <td>-5.337913</td>\n",
       "      <td>-5.725196</td>\n",
       "      <td>-0.387283</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>left</th>\n",
       "      <td>5</td>\n",
       "      <td>113</td>\n",
       "      <td>118</td>\n",
       "      <td>3.959533</td>\n",
       "      <td>4.888390</td>\n",
       "      <td>-6.028585</td>\n",
       "      <td>-6.462161</td>\n",
       "      <td>-0.433577</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>got</th>\n",
       "      <td>1</td>\n",
       "      <td>38</td>\n",
       "      <td>39</td>\n",
       "      <td>3.825232</td>\n",
       "      <td>5.071685</td>\n",
       "      <td>-7.576072</td>\n",
       "      <td>-8.047124</td>\n",
       "      <td>-0.471051</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>worked</th>\n",
       "      <td>12</td>\n",
       "      <td>266</td>\n",
       "      <td>278</td>\n",
       "      <td>4.772765</td>\n",
       "      <td>5.521923</td>\n",
       "      <td>-4.800779</td>\n",
       "      <td>-5.346684</td>\n",
       "      <td>-0.545905</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>said</th>\n",
       "      <td>3</td>\n",
       "      <td>92</td>\n",
       "      <td>95</td>\n",
       "      <td>3.673229</td>\n",
       "      <td>4.847966</td>\n",
       "      <td>-6.322316</td>\n",
       "      <td>-7.047124</td>\n",
       "      <td>-0.724808</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>studied</th>\n",
       "      <td>2</td>\n",
       "      <td>70</td>\n",
       "      <td>72</td>\n",
       "      <td>5.995157</td>\n",
       "      <td>5.998270</td>\n",
       "      <td>-6.711728</td>\n",
       "      <td>-7.462161</td>\n",
       "      <td>-0.750434</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>co</th>\n",
       "      <td>4</td>\n",
       "      <td>125</td>\n",
       "      <td>129</td>\n",
       "      <td>2.137176</td>\n",
       "      <td>3.130802</td>\n",
       "      <td>-5.884195</td>\n",
       "      <td>-6.725196</td>\n",
       "      <td>-0.841001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>created</th>\n",
       "      <td>2</td>\n",
       "      <td>100</td>\n",
       "      <td>102</td>\n",
       "      <td>3.602840</td>\n",
       "      <td>4.738686</td>\n",
       "      <td>-6.203263</td>\n",
       "      <td>-7.462161</td>\n",
       "      <td>-1.258898</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>then</th>\n",
       "      <td>1</td>\n",
       "      <td>79</td>\n",
       "      <td>80</td>\n",
       "      <td>1.825232</td>\n",
       "      <td>4.075071</td>\n",
       "      <td>-6.539547</td>\n",
       "      <td>-8.047124</td>\n",
       "      <td>-1.507577</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>founded</th>\n",
       "      <td>1</td>\n",
       "      <td>128</td>\n",
       "      <td>129</td>\n",
       "      <td>2.995157</td>\n",
       "      <td>4.603275</td>\n",
       "      <td>-5.850247</td>\n",
       "      <td>-8.047124</td>\n",
       "      <td>-2.196876</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           she+word count  he+word count  total count  PMI she+word  \\\n",
       "won                     3             56           59      4.772765   \n",
       "joined                  8            129          137      4.825232   \n",
       "started                 7            117          124      5.047625   \n",
       "continued               2             44           46      4.673229   \n",
       "called                  1             30           31      1.865874   \n",
       "developed               5             93           98      4.857654   \n",
       "has                    37            598          635      5.262096   \n",
       "helped                  2             49           51      4.535726   \n",
       "announced               1             33           34      2.825232   \n",
       "served                  4             85           89      4.602840   \n",
       "released                1             34           35      2.673229   \n",
       "is                     54            974         1028      4.469274   \n",
       "received                9            183          192      4.955629   \n",
       "left                    5            113          118      3.959533   \n",
       "got                     1             38           39      3.825232   \n",
       "worked                 12            266          278      4.772765   \n",
       "said                    3             92           95      3.673229   \n",
       "studied                 2             70           72      5.995157   \n",
       "co                      4            125          129      2.137176   \n",
       "created                 2            100          102      3.602840   \n",
       "then                    1             79           80      1.825232   \n",
       "founded                 1            128          129      2.995157   \n",
       "\n",
       "           PMI he+word  log prob of word | he  log prob of word | she  \\\n",
       "won           4.553110              -7.028585               -7.047124   \n",
       "joined        5.589872              -5.839107               -5.877199   \n",
       "started       5.226618              -5.978832               -6.047124   \n",
       "continued     5.030582              -7.369622               -7.462161   \n",
       "called        2.824946              -7.907278               -8.047124   \n",
       "developed     4.672099              -6.306886               -6.462161   \n",
       "has           5.159993              -3.635062               -3.799196   \n",
       "helped        5.050397              -7.217618               -7.462161   \n",
       "announced     3.274311              -7.774012               -8.047124   \n",
       "served        5.130055              -6.435210               -6.725196   \n",
       "released      2.961052              -7.732192               -8.047124   \n",
       "is            4.585574              -2.932216               -3.265764   \n",
       "received      5.504382              -5.337913               -5.725196   \n",
       "left          4.888390              -6.028585               -6.462161   \n",
       "got           5.071685              -7.576072               -8.047124   \n",
       "worked        5.521923              -4.800779               -5.346684   \n",
       "said          4.847966              -6.322316               -7.047124   \n",
       "studied       5.998270              -6.711728               -7.462161   \n",
       "co            3.130802              -5.884195               -6.725196   \n",
       "created       4.738686              -6.203263               -7.462161   \n",
       "then          4.075071              -6.539547               -8.047124   \n",
       "founded       4.603275              -5.850247               -8.047124   \n",
       "\n",
       "                LLR  \n",
       "won       -0.018539  \n",
       "joined    -0.038092  \n",
       "started   -0.068292  \n",
       "continued -0.092540  \n",
       "called    -0.139846  \n",
       "developed -0.155276  \n",
       "has       -0.164134  \n",
       "helped    -0.244543  \n",
       "announced -0.273112  \n",
       "served    -0.289986  \n",
       "released  -0.314932  \n",
       "is        -0.333548  \n",
       "received  -0.387283  \n",
       "left      -0.433577  \n",
       "got       -0.471051  \n",
       "worked    -0.545905  \n",
       "said      -0.724808  \n",
       "studied   -0.750434  \n",
       "co        -0.841001  \n",
       "created   -1.258898  \n",
       "then      -1.507577  \n",
       "founded   -2.196876  "
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Number of words in corpus of men and women's PROGRAMMING bios: \n",
    "n_women = sum(female_programming_u.values())\n",
    "n_men = sum(male_programming_u.values())\n",
    "\n",
    "# bigrams beginning with he or she, filtered by words appearing at least 30 times.\n",
    "she = collections.Counter({ k[1]: v for k, v in female_programming.items() if v >= 30 and k[0] == \"she\" })\n",
    "he = collections.Counter({ k[1]: v for k, v in male_programming.items() if v >= 30 and k[0] == \"he\" })\n",
    "\n",
    "he_count = male_programming_u['he']\n",
    "she_count = female_programming_u['she']\n",
    "\n",
    "for word in he:\n",
    "    if word not in she:\n",
    "        she[word] = female_programming[(\"she\", word)]\n",
    "for word in she:\n",
    "    if word not in he:\n",
    "        he[word] = male_programming[(\"he\", word)]\n",
    "\n",
    "columns = ['Word following he/she', \"she+word count\"]\n",
    "df = pd.DataFrame([(k,v) for k,v in she.items()], columns=columns)\n",
    "df.set_index(['Word following he/she'], inplace=True)\n",
    "columns2 = ['Word following he/she', \"he+word count\",]\n",
    "df2 = pd.DataFrame([(k,v) for k,v in he.items()], columns=columns2)\n",
    "df2.set_index(['Word following he/she'], inplace=True)\n",
    "\n",
    "df = pd.concat([df,df2], axis=1)\n",
    "df['total count'] = df.fillna(0)[\"she+word count\"] + df.fillna(0)[\"he+word count\"]\n",
    "df.sort_values(by=['total count'], ascending=False, inplace=True)\n",
    "\n",
    "df['PMI she+word'] = df.apply(lambda row: pmi(she_count, female_programming_u[row.name], \n",
    "                                              row[\"she+word count\"], n_women), axis=1)\n",
    "df['PMI he+word'] = df.apply(lambda row: pmi(he_count, male_programming_u[row.name], \n",
    "                                             row[\"he+word count\"], n_men), axis=1)\n",
    "\n",
    "# Log probabilities of word given he/she. Add 1 for smoothing\n",
    "df['log prob of word | he'] = np.log2(\n",
    "    (df.fillna(0)[\"he+word count\"] + 1)) - np.log2((sum(he.values()) + 1))\n",
    "df['log prob of word | she'] = np.log2(\n",
    "    (df.fillna(0)[\"she+word count\"] + 1)) - np.log2((sum(she.values()) + 1))\n",
    "\n",
    "# Log likelihood ratio she : he\n",
    "df['LLR'] = df['log prob of word | she'] - df['log prob of word | he']\n",
    "df.sort_values(by=['LLR'], ascending=False, inplace=True)\n",
    "\n",
    "he_words = df.loc[(df['PMI he+word'] > 0) & (df['LLR'] < 0), ]\n",
    "he_words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "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>she+word count</th>\n",
       "      <th>he+word count</th>\n",
       "      <th>total count</th>\n",
       "      <th>PMI she+word</th>\n",
       "      <th>PMI he+word</th>\n",
       "      <th>log prob of word | he</th>\n",
       "      <th>log prob of word | she</th>\n",
       "      <th>LLR</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>did</th>\n",
       "      <td>13</td>\n",
       "      <td>71</td>\n",
       "      <td>84</td>\n",
       "      <td>5.269332</td>\n",
       "      <td>4.626854</td>\n",
       "      <td>-6.691550</td>\n",
       "      <td>-5.239769</td>\n",
       "      <td>1.451781</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>died</th>\n",
       "      <td>5</td>\n",
       "      <td>30</td>\n",
       "      <td>35</td>\n",
       "      <td>4.732123</td>\n",
       "      <td>4.993683</td>\n",
       "      <td>-7.907278</td>\n",
       "      <td>-6.462161</td>\n",
       "      <td>1.445117</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>led</th>\n",
       "      <td>6</td>\n",
       "      <td>38</td>\n",
       "      <td>44</td>\n",
       "      <td>4.722139</td>\n",
       "      <td>3.976907</td>\n",
       "      <td>-7.576072</td>\n",
       "      <td>-6.239769</td>\n",
       "      <td>1.336303</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>began</th>\n",
       "      <td>10</td>\n",
       "      <td>79</td>\n",
       "      <td>89</td>\n",
       "      <td>5.362889</td>\n",
       "      <td>4.819930</td>\n",
       "      <td>-6.539547</td>\n",
       "      <td>-5.587692</td>\n",
       "      <td>0.951854</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>holds</th>\n",
       "      <td>3</td>\n",
       "      <td>30</td>\n",
       "      <td>33</td>\n",
       "      <td>5.995157</td>\n",
       "      <td>5.225008</td>\n",
       "      <td>-7.907278</td>\n",
       "      <td>-7.047124</td>\n",
       "      <td>0.860154</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>took</th>\n",
       "      <td>7</td>\n",
       "      <td>63</td>\n",
       "      <td>70</td>\n",
       "      <td>5.632587</td>\n",
       "      <td>5.135263</td>\n",
       "      <td>-6.861475</td>\n",
       "      <td>-6.047124</td>\n",
       "      <td>0.814351</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>met</th>\n",
       "      <td>3</td>\n",
       "      <td>31</td>\n",
       "      <td>34</td>\n",
       "      <td>4.995157</td>\n",
       "      <td>5.051715</td>\n",
       "      <td>-7.861475</td>\n",
       "      <td>-7.047124</td>\n",
       "      <td>0.814351</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>could</th>\n",
       "      <td>4</td>\n",
       "      <td>41</td>\n",
       "      <td>45</td>\n",
       "      <td>3.568892</td>\n",
       "      <td>3.910135</td>\n",
       "      <td>-7.469157</td>\n",
       "      <td>-6.725196</td>\n",
       "      <td>0.743961</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>had</th>\n",
       "      <td>32</td>\n",
       "      <td>282</td>\n",
       "      <td>314</td>\n",
       "      <td>4.865874</td>\n",
       "      <td>4.738057</td>\n",
       "      <td>-4.716816</td>\n",
       "      <td>-4.002730</td>\n",
       "      <td>0.714087</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>wrote</th>\n",
       "      <td>14</td>\n",
       "      <td>137</td>\n",
       "      <td>151</td>\n",
       "      <td>5.444960</td>\n",
       "      <td>5.301638</td>\n",
       "      <td>-5.752950</td>\n",
       "      <td>-5.140233</td>\n",
       "      <td>0.612717</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>earned</th>\n",
       "      <td>4</td>\n",
       "      <td>45</td>\n",
       "      <td>49</td>\n",
       "      <td>6.410195</td>\n",
       "      <td>5.446007</td>\n",
       "      <td>-7.337913</td>\n",
       "      <td>-6.725196</td>\n",
       "      <td>0.612717</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>spent</th>\n",
       "      <td>4</td>\n",
       "      <td>45</td>\n",
       "      <td>49</td>\n",
       "      <td>5.187802</td>\n",
       "      <td>5.417056</td>\n",
       "      <td>-7.337913</td>\n",
       "      <td>-6.725196</td>\n",
       "      <td>0.612717</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>used</th>\n",
       "      <td>3</td>\n",
       "      <td>36</td>\n",
       "      <td>39</td>\n",
       "      <td>3.187802</td>\n",
       "      <td>3.083234</td>\n",
       "      <td>-7.652021</td>\n",
       "      <td>-7.047124</td>\n",
       "      <td>0.604897</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>later</th>\n",
       "      <td>7</td>\n",
       "      <td>74</td>\n",
       "      <td>81</td>\n",
       "      <td>3.758118</td>\n",
       "      <td>3.507328</td>\n",
       "      <td>-6.632656</td>\n",
       "      <td>-6.047124</td>\n",
       "      <td>0.585532</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>went</th>\n",
       "      <td>4</td>\n",
       "      <td>47</td>\n",
       "      <td>51</td>\n",
       "      <td>4.471595</td>\n",
       "      <td>4.568449</td>\n",
       "      <td>-7.276512</td>\n",
       "      <td>-6.725196</td>\n",
       "      <td>0.551316</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>made</th>\n",
       "      <td>3</td>\n",
       "      <td>38</td>\n",
       "      <td>41</td>\n",
       "      <td>3.258192</td>\n",
       "      <td>3.414356</td>\n",
       "      <td>-7.576072</td>\n",
       "      <td>-7.047124</td>\n",
       "      <td>0.528949</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>wanted</th>\n",
       "      <td>2</td>\n",
       "      <td>31</td>\n",
       "      <td>33</td>\n",
       "      <td>5.187802</td>\n",
       "      <td>4.814023</td>\n",
       "      <td>-7.861475</td>\n",
       "      <td>-7.462161</td>\n",
       "      <td>0.399313</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>was</th>\n",
       "      <td>150</td>\n",
       "      <td>1613</td>\n",
       "      <td>1763</td>\n",
       "      <td>4.855469</td>\n",
       "      <td>4.797291</td>\n",
       "      <td>-2.205050</td>\n",
       "      <td>-1.808719</td>\n",
       "      <td>0.396331</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>attended</th>\n",
       "      <td>6</td>\n",
       "      <td>79</td>\n",
       "      <td>85</td>\n",
       "      <td>5.580120</td>\n",
       "      <td>5.609407</td>\n",
       "      <td>-6.539547</td>\n",
       "      <td>-6.239769</td>\n",
       "      <td>0.299778</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>became</th>\n",
       "      <td>13</td>\n",
       "      <td>160</td>\n",
       "      <td>173</td>\n",
       "      <td>4.718317</td>\n",
       "      <td>4.896821</td>\n",
       "      <td>-5.530558</td>\n",
       "      <td>-5.239769</td>\n",
       "      <td>0.290789</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>graduated</th>\n",
       "      <td>7</td>\n",
       "      <td>91</td>\n",
       "      <td>98</td>\n",
       "      <td>6.217550</td>\n",
       "      <td>6.020351</td>\n",
       "      <td>-6.337913</td>\n",
       "      <td>-6.047124</td>\n",
       "      <td>0.290789</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>designed</th>\n",
       "      <td>2</td>\n",
       "      <td>36</td>\n",
       "      <td>38</td>\n",
       "      <td>3.825232</td>\n",
       "      <td>4.289139</td>\n",
       "      <td>-7.652021</td>\n",
       "      <td>-7.462161</td>\n",
       "      <td>0.189860</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>lives</th>\n",
       "      <td>2</td>\n",
       "      <td>37</td>\n",
       "      <td>39</td>\n",
       "      <td>3.747230</td>\n",
       "      <td>4.946796</td>\n",
       "      <td>-7.613547</td>\n",
       "      <td>-7.462161</td>\n",
       "      <td>0.151386</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>would</th>\n",
       "      <td>12</td>\n",
       "      <td>167</td>\n",
       "      <td>179</td>\n",
       "      <td>4.936263</td>\n",
       "      <td>4.741568</td>\n",
       "      <td>-5.469157</td>\n",
       "      <td>-5.346684</td>\n",
       "      <td>0.122473</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>also</th>\n",
       "      <td>29</td>\n",
       "      <td>410</td>\n",
       "      <td>439</td>\n",
       "      <td>4.830770</td>\n",
       "      <td>4.728695</td>\n",
       "      <td>-4.178480</td>\n",
       "      <td>-4.140233</td>\n",
       "      <td>0.038247</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>currently</th>\n",
       "      <td>2</td>\n",
       "      <td>41</td>\n",
       "      <td>43</td>\n",
       "      <td>4.673229</td>\n",
       "      <td>4.326403</td>\n",
       "      <td>-7.469157</td>\n",
       "      <td>-7.462161</td>\n",
       "      <td>0.006996</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>moved</th>\n",
       "      <td>4</td>\n",
       "      <td>69</td>\n",
       "      <td>73</td>\n",
       "      <td>5.294717</td>\n",
       "      <td>5.179375</td>\n",
       "      <td>-6.732192</td>\n",
       "      <td>-6.725196</td>\n",
       "      <td>0.006996</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           she+word count  he+word count  total count  PMI she+word  \\\n",
       "did                    13             71           84      5.269332   \n",
       "died                    5             30           35      4.732123   \n",
       "led                     6             38           44      4.722139   \n",
       "began                  10             79           89      5.362889   \n",
       "holds                   3             30           33      5.995157   \n",
       "took                    7             63           70      5.632587   \n",
       "met                     3             31           34      4.995157   \n",
       "could                   4             41           45      3.568892   \n",
       "had                    32            282          314      4.865874   \n",
       "wrote                  14            137          151      5.444960   \n",
       "earned                  4             45           49      6.410195   \n",
       "spent                   4             45           49      5.187802   \n",
       "used                    3             36           39      3.187802   \n",
       "later                   7             74           81      3.758118   \n",
       "went                    4             47           51      4.471595   \n",
       "made                    3             38           41      3.258192   \n",
       "wanted                  2             31           33      5.187802   \n",
       "was                   150           1613         1763      4.855469   \n",
       "attended                6             79           85      5.580120   \n",
       "became                 13            160          173      4.718317   \n",
       "graduated               7             91           98      6.217550   \n",
       "designed                2             36           38      3.825232   \n",
       "lives                   2             37           39      3.747230   \n",
       "would                  12            167          179      4.936263   \n",
       "also                   29            410          439      4.830770   \n",
       "currently               2             41           43      4.673229   \n",
       "moved                   4             69           73      5.294717   \n",
       "\n",
       "           PMI he+word  log prob of word | he  log prob of word | she  \\\n",
       "did           4.626854              -6.691550               -5.239769   \n",
       "died          4.993683              -7.907278               -6.462161   \n",
       "led           3.976907              -7.576072               -6.239769   \n",
       "began         4.819930              -6.539547               -5.587692   \n",
       "holds         5.225008              -7.907278               -7.047124   \n",
       "took          5.135263              -6.861475               -6.047124   \n",
       "met           5.051715              -7.861475               -7.047124   \n",
       "could         3.910135              -7.469157               -6.725196   \n",
       "had           4.738057              -4.716816               -4.002730   \n",
       "wrote         5.301638              -5.752950               -5.140233   \n",
       "earned        5.446007              -7.337913               -6.725196   \n",
       "spent         5.417056              -7.337913               -6.725196   \n",
       "used          3.083234              -7.652021               -7.047124   \n",
       "later         3.507328              -6.632656               -6.047124   \n",
       "went          4.568449              -7.276512               -6.725196   \n",
       "made          3.414356              -7.576072               -7.047124   \n",
       "wanted        4.814023              -7.861475               -7.462161   \n",
       "was           4.797291              -2.205050               -1.808719   \n",
       "attended      5.609407              -6.539547               -6.239769   \n",
       "became        4.896821              -5.530558               -5.239769   \n",
       "graduated     6.020351              -6.337913               -6.047124   \n",
       "designed      4.289139              -7.652021               -7.462161   \n",
       "lives         4.946796              -7.613547               -7.462161   \n",
       "would         4.741568              -5.469157               -5.346684   \n",
       "also          4.728695              -4.178480               -4.140233   \n",
       "currently     4.326403              -7.469157               -7.462161   \n",
       "moved         5.179375              -6.732192               -6.725196   \n",
       "\n",
       "                LLR  \n",
       "did        1.451781  \n",
       "died       1.445117  \n",
       "led        1.336303  \n",
       "began      0.951854  \n",
       "holds      0.860154  \n",
       "took       0.814351  \n",
       "met        0.814351  \n",
       "could      0.743961  \n",
       "had        0.714087  \n",
       "wrote      0.612717  \n",
       "earned     0.612717  \n",
       "spent      0.612717  \n",
       "used       0.604897  \n",
       "later      0.585532  \n",
       "went       0.551316  \n",
       "made       0.528949  \n",
       "wanted     0.399313  \n",
       "was        0.396331  \n",
       "attended   0.299778  \n",
       "became     0.290789  \n",
       "graduated  0.290789  \n",
       "designed   0.189860  \n",
       "lives      0.151386  \n",
       "would      0.122473  \n",
       "also       0.038247  \n",
       "currently  0.006996  \n",
       "moved      0.006996  "
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "she_words = df.loc[(df['PMI she+word'] > 0) & (df['LLR'] > 0), ]\n",
    "she_words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Pretty low word counts ... probably shouldn't use programming?"
   ]
  },
  {
   "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": "3.6.5"
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}