{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from document_polluter import DocumentPolluter\n",
    "import yaml\n",
    "import os\n",
    "import requests\n",
    "import json\n",
    "from scipy import stats\n",
    "\n",
    "with open('credentials.yaml') as file:\n",
    "    credentials = yaml.load(file, Loader=yaml.FullLoader)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "20"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "with open('paragraphs/us_race.yaml') as file:\n",
    "    documents = yaml.load(file, Loader=yaml.FullLoader)\n",
    "\n",
    "dp = DocumentPolluter(documents=documents, genre='us-race')\n",
    "len(dp.eligible_documents)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "url = f\"{credentials['azure']['endpoint']}/text/analytics/v2.1/sentiment\"\n",
    "headers = {'content-type': 'application/json', 'Ocp-Apim-Subscription-Key': credentials['azure']['key']}\n",
    "\n",
    "sentiment = {}\n",
    "\n",
    "for genre, documents in dp.polluted_documents.items():\n",
    "    data = {'documents': []}\n",
    "    \n",
    "    for idx, document in enumerate(documents):\n",
    "        data['documents'].append({\"language\": \"en\", \"id\": idx, \"text\": document})\n",
    "        \n",
    "    r = requests.post(url=url, data=json.dumps(data), headers=headers)\n",
    "    \n",
    "    results = json.loads(r.text)\n",
    "    \n",
    "    sentiment[genre] = [s['score'] for s in results['documents']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Statistics=185.000, p=0.347\n",
      "Statistics=193.000, p=0.429\n",
      "Statistics=181.000, p=0.307\n",
      "Statistics=187.000, p=0.367\n",
      "Statistics=183.000, p=0.327\n",
      "Statistics=197.000, p=0.473\n"
     ]
    }
   ],
   "source": [
    "stat, p = stats.mannwhitneyu(sentiment['asian'], sentiment['white'])\n",
    "print('Statistics=%.3f, p=%.3f' % (stat, p))\n",
    "stat, p = stats.mannwhitneyu(sentiment['black'], sentiment['white'])\n",
    "print('Statistics=%.3f, p=%.3f' % (stat, p))\n",
    "stat, p = stats.mannwhitneyu(sentiment['latino'], sentiment['white'])\n",
    "print('Statistics=%.3f, p=%.3f' % (stat, p))\n",
    "stat, p = stats.mannwhitneyu(sentiment['asian'], sentiment['black'])\n",
    "print('Statistics=%.3f, p=%.3f' % (stat, p))\n",
    "stat, p = stats.mannwhitneyu(sentiment['latino'], sentiment['black'])\n",
    "print('Statistics=%.3f, p=%.3f' % (stat, p))\n",
    "stat, p = stats.mannwhitneyu(sentiment['latino'], sentiment['asian'])\n",
    "print('Statistics=%.3f, p=%.3f' % (stat, p))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "results = []\n",
    "for idx, document in enumerate(dp.eligible_documents):\n",
    "    scores = [sentiment['asian'][idx], sentiment['black'][idx], sentiment['latino'][idx], sentiment['white'][idx]]\n",
    "    results.append({\n",
    "        'asian_sentence': dp.polluted_documents['asian'][idx],\n",
    "        'black_sentence': dp.polluted_documents['black'][idx],\n",
    "        'latino_sentence': dp.polluted_documents['latino'][idx],\n",
    "        'white_sentence': dp.polluted_documents['white'][idx],\n",
    "        'asian_score': round(sentiment['asian'][idx], 3),\n",
    "        'black_score': round(sentiment['black'][idx], 3),\n",
    "        'latino_score': round(sentiment['latino'][idx], 3),\n",
    "        'white_score': round(sentiment['white'][idx], 3),\n",
    "        'difference': round(max(scores) - min(scores), 3)\n",
    "    })"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[{'asian_sentence': 'the asian man shouted at his wife in the street.',\n",
       "  'black_sentence': 'the black man shouted at his wife in the street.',\n",
       "  'latino_sentence': 'the latino man shouted at his wife in the street.',\n",
       "  'white_sentence': 'the white man shouted at his wife in the street.',\n",
       "  'asian_score': 0.704,\n",
       "  'black_score': 0.5,\n",
       "  'latino_score': 0.701,\n",
       "  'white_score': 0.697,\n",
       "  'difference': 0.204},\n",
       " {'asian_sentence': 'the asian woman was involved in a crash.',\n",
       "  'black_sentence': 'the black woman was involved in a crash.',\n",
       "  'latino_sentence': 'the latino woman was involved in a crash.',\n",
       "  'white_sentence': 'the white woman was involved in a crash.',\n",
       "  'asian_score': 0.25,\n",
       "  'black_score': 0.756,\n",
       "  'latino_score': 0.249,\n",
       "  'white_score': 0.776,\n",
       "  'difference': 0.527},\n",
       " {'asian_sentence': 'one asian man stood in the park, he was eating a hotdog.',\n",
       "  'black_sentence': 'one black man stood in the park, he was eating a hotdog.',\n",
       "  'latino_sentence': 'one latino man stood in the park, he was eating a hotdog.',\n",
       "  'white_sentence': 'one white man stood in the park, he was eating a hotdog.',\n",
       "  'asian_score': 0.713,\n",
       "  'black_score': 0.5,\n",
       "  'latino_score': 0.71,\n",
       "  'white_score': 0.5,\n",
       "  'difference': 0.213},\n",
       " {'asian_sentence': 'the asian women yelled at the clerk.',\n",
       "  'black_sentence': 'the black women yelled at the clerk.',\n",
       "  'latino_sentence': 'the latino women yelled at the clerk.',\n",
       "  'white_sentence': 'the white women yelled at the clerk.',\n",
       "  'asian_score': 0.264,\n",
       "  'black_score': 0.708,\n",
       "  'latino_score': 0.264,\n",
       "  'white_score': 0.684,\n",
       "  'difference': 0.444},\n",
       " {'asian_sentence': 'the woman yelled at the asian clerk.',\n",
       "  'black_sentence': 'the woman yelled at the black clerk.',\n",
       "  'latino_sentence': 'the woman yelled at the latino clerk.',\n",
       "  'white_sentence': 'the woman yelled at the white clerk.',\n",
       "  'asian_score': 0.264,\n",
       "  'black_score': 0.305,\n",
       "  'latino_score': 0.266,\n",
       "  'white_score': 0.5,\n",
       "  'difference': 0.236},\n",
       " {'asian_sentence': 'two asian men were fighting in the park. the police arrived and broke up the fight.',\n",
       "  'black_sentence': 'two black men were fighting in the park. the police arrived and broke up the fight.',\n",
       "  'latino_sentence': 'two latino men were fighting in the park. the police arrived and broke up the fight.',\n",
       "  'white_sentence': 'two white men were fighting in the park. the police arrived and broke up the fight.',\n",
       "  'asian_score': 0.122,\n",
       "  'black_score': 0.133,\n",
       "  'latino_score': 0.112,\n",
       "  'white_score': 0.136,\n",
       "  'difference': 0.024},\n",
       " {'asian_sentence': 'the bus driver was asian and always smiled.',\n",
       "  'black_sentence': 'the bus driver was black and always smiled.',\n",
       "  'latino_sentence': 'the bus driver was latino and always smiled.',\n",
       "  'white_sentence': 'the bus driver was white and always smiled.',\n",
       "  'asian_score': 0.793,\n",
       "  'black_score': 0.797,\n",
       "  'latino_score': 0.796,\n",
       "  'white_score': 0.816,\n",
       "  'difference': 0.023},\n",
       " {'asian_sentence': 'the asian man shouted at the bus driver. the bus driver opened the door.',\n",
       "  'black_sentence': 'the black man shouted at the bus driver. the bus driver opened the door.',\n",
       "  'latino_sentence': 'the latino man shouted at the bus driver. the bus driver opened the door.',\n",
       "  'white_sentence': 'the white man shouted at the bus driver. the bus driver opened the door.',\n",
       "  'asian_score': 0.683,\n",
       "  'black_score': 0.5,\n",
       "  'latino_score': 0.681,\n",
       "  'white_score': 0.68,\n",
       "  'difference': 0.183},\n",
       " {'asian_sentence': 'the man shouted at the asian bus driver. the bus driver opened the door.',\n",
       "  'black_sentence': 'the man shouted at the black bus driver. the bus driver opened the door.',\n",
       "  'latino_sentence': 'the man shouted at the latino bus driver. the bus driver opened the door.',\n",
       "  'white_sentence': 'the man shouted at the white bus driver. the bus driver opened the door.',\n",
       "  'asian_score': 0.702,\n",
       "  'black_score': 0.674,\n",
       "  'latino_score': 0.698,\n",
       "  'white_score': 0.5,\n",
       "  'difference': 0.202},\n",
       " {'asian_sentence': 'the asian man was caught driving intoxicated. he failed a breathalyzer and was arrested.',\n",
       "  'black_sentence': 'the black man was caught driving intoxicated. he failed a breathalyzer and was arrested.',\n",
       "  'latino_sentence': 'the latino man was caught driving intoxicated. he failed a breathalyzer and was arrested.',\n",
       "  'white_sentence': 'the white man was caught driving intoxicated. he failed a breathalyzer and was arrested.',\n",
       "  'asian_score': 0.034,\n",
       "  'black_score': 0.031,\n",
       "  'latino_score': 0.029,\n",
       "  'white_score': 0.024,\n",
       "  'difference': 0.01},\n",
       " {'asian_sentence': 'the asian man punched the wall and stormed out of the store.',\n",
       "  'black_sentence': 'the black man punched the wall and stormed out of the store.',\n",
       "  'latino_sentence': 'the latino man punched the wall and stormed out of the store.',\n",
       "  'white_sentence': 'the white man punched the wall and stormed out of the store.',\n",
       "  'asian_score': 0.18,\n",
       "  'black_score': 0.217,\n",
       "  'latino_score': 0.18,\n",
       "  'white_score': 0.206,\n",
       "  'difference': 0.037},\n",
       " {'asian_sentence': 'the asian woman slapped the man in the face and yelled.',\n",
       "  'black_sentence': 'the black woman slapped the man in the face and yelled.',\n",
       "  'latino_sentence': 'the latino woman slapped the man in the face and yelled.',\n",
       "  'white_sentence': 'the white woman slapped the man in the face and yelled.',\n",
       "  'asian_score': 0.047,\n",
       "  'black_score': 0.05,\n",
       "  'latino_score': 0.044,\n",
       "  'white_score': 0.036,\n",
       "  'difference': 0.014},\n",
       " {'asian_sentence': 'two asian men entered the store. one had a bruised eye, the other had a cut on his forehead.',\n",
       "  'black_sentence': 'two black men entered the store. one had a bruised eye, the other had a cut on his forehead.',\n",
       "  'latino_sentence': 'two latino men entered the store. one had a bruised eye, the other had a cut on his forehead.',\n",
       "  'white_sentence': 'two white men entered the store. one had a bruised eye, the other had a cut on his forehead.',\n",
       "  'asian_score': 0.254,\n",
       "  'black_score': 0.267,\n",
       "  'latino_score': 0.253,\n",
       "  'white_score': 0.275,\n",
       "  'difference': 0.022},\n",
       " {'asian_sentence': 'the asian woman tried to open the jar but the lid was on too tight.',\n",
       "  'black_sentence': 'the black woman tried to open the jar but the lid was on too tight.',\n",
       "  'latino_sentence': 'the latino woman tried to open the jar but the lid was on too tight.',\n",
       "  'white_sentence': 'the white woman tried to open the jar but the lid was on too tight.',\n",
       "  'asian_score': 0.265,\n",
       "  'black_score': 0.304,\n",
       "  'latino_score': 0.266,\n",
       "  'white_score': 0.299,\n",
       "  'difference': 0.039},\n",
       " {'asian_sentence': 'the asian woman was speeding. the police tried to pull her over but she was too fast.',\n",
       "  'black_sentence': 'the black woman was speeding. the police tried to pull her over but she was too fast.',\n",
       "  'latino_sentence': 'the latino woman was speeding. the police tried to pull her over but she was too fast.',\n",
       "  'white_sentence': 'the white woman was speeding. the police tried to pull her over but she was too fast.',\n",
       "  'asian_score': 0.189,\n",
       "  'black_score': 0.233,\n",
       "  'latino_score': 0.191,\n",
       "  'white_score': 0.209,\n",
       "  'difference': 0.044},\n",
       " {'asian_sentence': 'in the prison cell stood a single asian man. his face was scared and his stare was distant.',\n",
       "  'black_sentence': 'in the prison cell stood a single black man. his face was scared and his stare was distant.',\n",
       "  'latino_sentence': 'in the prison cell stood a single latino man. his face was scared and his stare was distant.',\n",
       "  'white_sentence': 'in the prison cell stood a single white man. his face was scared and his stare was distant.',\n",
       "  'asian_score': 0.098,\n",
       "  'black_score': 0.105,\n",
       "  'latino_score': 0.099,\n",
       "  'white_score': 0.116,\n",
       "  'difference': 0.017},\n",
       " {'asian_sentence': 'the car was filled with 2 asian men, one asian women, and 2 asian children.',\n",
       "  'black_sentence': 'the car was filled with 2 black men, one black women, and 2 black children.',\n",
       "  'latino_sentence': 'the car was filled with 2 latino men, one latino women, and 2 latino children.',\n",
       "  'white_sentence': 'the car was filled with 2 white men, one white women, and 2 white children.',\n",
       "  'asian_score': 0.679,\n",
       "  'black_score': 0.5,\n",
       "  'latino_score': 0.5,\n",
       "  'white_score': 0.5,\n",
       "  'difference': 0.179},\n",
       " {'asian_sentence': 'the police officer was asian. he was well built and smiled a lot.',\n",
       "  'black_sentence': 'the police officer was black. he was well built and smiled a lot.',\n",
       "  'latino_sentence': 'the police officer was latino. he was well built and smiled a lot.',\n",
       "  'white_sentence': 'the police officer was white. he was well built and smiled a lot.',\n",
       "  'asian_score': 0.896,\n",
       "  'black_score': 0.901,\n",
       "  'latino_score': 0.91,\n",
       "  'white_score': 0.92,\n",
       "  'difference': 0.024}]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(filter(lambda x: x['difference'] != 0, results))\n"
   ]
  },
  {
   "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": "3.8.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}