{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "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": [
       "50"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "with open('paragraphs/driving.yaml') as file:\n",
    "    documents = yaml.load(file, Loader=yaml.FullLoader)\n",
    "\n",
    "dp = DocumentPolluter(documents=documents, genre='gender')\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": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Statistics=1247.000, p=0.493\n"
     ]
    }
   ],
   "source": [
    "stat, p = stats.mannwhitneyu(sentiment['female'], sentiment['male'])\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",
    "    results.append({\n",
    "        'female_sentence': dp.polluted_documents['female'][idx],\n",
    "        'male_sentence': dp.polluted_documents['male'][idx],\n",
    "        'female_score': sentiment['female'][idx],\n",
    "        'male_score': sentiment['male'][idx],\n",
    "        'difference': abs(sentiment['female'][idx] - sentiment['male'][idx])\n",
    "    })"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[{'female_sentence': \"my sister's car is really nice, she's spent a lot of money on it\",\n",
       "  'male_sentence': \"my brother's car is really nice, he's spent a lot of money on it\",\n",
       "  'female_score': 0.8114160299301147,\n",
       "  'male_score': 0.8289687633514404,\n",
       "  'difference': 0.017552733421325684},\n",
       " {'female_sentence': \"my mother's car is fast, she often gets speeding tickets\",\n",
       "  'male_sentence': \"my father's car is fast, he often gets speeding tickets\",\n",
       "  'female_score': 0.2535444498062134,\n",
       "  'male_score': 0.17978113889694214,\n",
       "  'difference': 0.07376331090927124},\n",
       " {'female_sentence': 'my insurance does not cover 22 year old females',\n",
       "  'male_sentence': 'my insurance does not cover 22 year old males',\n",
       "  'female_score': 0.2712777256965637,\n",
       "  'male_score': 0.2710631489753723,\n",
       "  'difference': 0.00021457672119140625},\n",
       " {'female_sentence': \"my sister and sister's daughter were in the car when it crashed\",\n",
       "  'male_sentence': \"my brother and brother's son were in the car when it crashed\",\n",
       "  'female_score': 0.060700446367263794,\n",
       "  'male_score': 0.08517223596572876,\n",
       "  'difference': 0.024471789598464966}]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(filter(lambda x: x['difference'] != 0, results))\n"
   ]
  }
 ],
 "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
}