preprocessing_2.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "from nltk.sentiment.vader import SentimentIntensityAnalyzer\n",
    "from nrclex import NRCLex\n",
    "with open('conver.json', 'r') as json_file:\n",
    "    data = json.load(json_file)\n",
    "analyzer = SentimentIntensityAnalyzer()\n",
    "\n",
    "def analyze_emotions(conversation,word_freq):\n",
    "    nrc_result = NRCLex(conversation)\n",
    "    emotio = nrc_result.affect_frequencies\n",
    "    emotion_words = nrc_result.affect_dict\n",
    "    emotion_categories = {\n",
    "        \"fear\": [],\n",
    "        \"anger\": [],\n",
    "        \"anticip\": [],\n",
    "        \"trust\": [],\n",
    "        \"surprise\": [],\n",
    "        \"positive\": [],\n",
    "        \"negative\": [],\n",
    "        \"sadness\": [],\n",
    "        \"disgust\": [],\n",
    "        \"joy\": [],\n",
    "        \"anticipation\": []\n",
    "    }\n",
    "    for word, emotions in emotion_words.items():\n",
    "        for emotion in emotions:\n",
    "            for category in emotion_categories.keys():\n",
    "                if category in emotion:\n",
    "                    if word in word_freq:\n",
    "                        frequency = word_freq[word]\n",
    "                        emotion_categories[category].append((word,frequency))\n",
    "\n",
    "    \n",
    "    result = {\n",
    "        \"emotions\": emotio,\n",
    "        \"emotion_words\": emotion_categories\n",
    "    }\n",
    "    \n",
    "    return result\n",
    "result_data = []\n",
    "\n",
    "for item in data:\n",
    "    sender = item[\"sender\"]\n",
    "    receiver_list = item[\"receiverList\"]\n",
    "    \n",
    "    receiver_results = []\n",
    "    \n",
    "    for receiver_item in receiver_list:\n",
    "        conversation = receiver_item[\" conversation\"]\n",
    "        from collections import Counter\n",
    "        import re\n",
    "        conversation = receiver_item[\" conversation\"]\n",
    "        words = re.findall(r'\\w+', conversation.lower())  \n",
    "        word_freq = Counter(words)\n",
    "        emotions_result = analyze_emotions(conversation,word_freq)\n",
    "        \n",
    "        receiver_result = {\n",
    "            \"receiver\": receiver_item[\"receiver\"],\n",
    "            \"conversation_result\": emotions_result\n",
    "        }\n",
    "        \n",
    "        receiver_results.append(receiver_result)\n",
    "    \n",
    "    sender_result = {\n",
    "        \"sender\": sender,\n",
    "        \"receiverList_results\": receiver_results\n",
    "    }\n",
    "    \n",
    "    result_data.append(sender_result)\n",
    "with open('conver1.json', 'w') as json_file:\n",
    "    json.dump(result_data,json_file, indent=4)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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": "3.11.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}