Sentiment-Analysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#Reading all the files (stop words, negative and positive words into 3 different dictionaries)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Populating the interactive namespace from numpy and matplotlib\n"
     ]
    }
   ],
   "source": [
    "from IPython.core.interactiveshell import InteractiveShell\n",
    "InteractiveShell.ast_node_interactivity = \"all\"\n",
    "\n",
    "import seaborn as sns\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "%pylab inline\n",
    "\n",
    "import sklearn as sk\n",
    "import sklearn.tree as tree\n",
    "from IPython.display import Image  \n",
    "\n",
    "import bs4 as bs\n",
    "import csv\n",
    "import requests\n",
    "import urllib\n",
    "from urllib.request import urlopen"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Sentiment Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "##Reading stop words into stop_words dictionary\n",
    "\n",
    "stop_words={}\n",
    "with open(\"C:\\\\Users\\\\shikh\\\\Desktop\\\\Practicum\\\\stop_words.txt\",'r') as sw:    ##opening stop_words.txt file\n",
    "    for line in sw:                        \n",
    "        line =line.strip()               ##stripping any blank spaces from the beginning & end of line\n",
    "        stop_words[line] = 0             ## Entering the word to stop_words dictionary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "##Reading negative words into negative_words dictionary\n",
    "negative_words={}\n",
    "with open(\"C:\\\\Users\\\\shikh\\\\Desktop\\\\Practicum\\\\negative-words.txt\",'r') as nw:          ##opening negative_words.txt file\n",
    "    for line in nw:\n",
    "        line = line.strip()                         ##stripping any blank spaces from the beginning & end of line\n",
    "        negative_words[line] = 0                     ## Entering the word to negative_words dictionary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "##Reading negative words into positive_words dictionary\n",
    "positive_words={}\n",
    "with open(\"C:\\\\Users\\\\shikh\\\\Desktop\\\\Practicum\\\\positive-words.txt\",'r') as pw:   ##opening postive_words.txt file\n",
    "    for line in pw:\n",
    "        line = line.strip()                  ##stripping any blank spaces from the beginning & end of line\n",
    "        positive_words[line] = 0             ## Entering the word to postive_words dictionary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "119\n",
      "4783\n",
      "2007\n"
     ]
    }
   ],
   "source": [
    "# Check if words have been loaded or not\n",
    "print(len(stop_words))                  \n",
    "print(len(negative_words))\n",
    "print(len(positive_words))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "bloomberg:\n",
      "apple\n",
      "loses\n",
      "-\n",
      "key\n",
      "execs\n",
      "in\n",
      "india\n",
      "struggles\n",
      "-\n",
      "jul.\n",
      "16,\n",
      "2018\n",
      "8:05\n",
      "am\n",
      "et|about:\n",
      "apple\n",
      "inc.\n",
      "aapl|by:\n",
      "brandy\n",
      "betz,\n",
      "sa\n",
      "news\n",
      "editor\n",
      "apple\n",
      "nasdaq:aapl\n",
      "has\n",
      "lost\n",
      "-\n",
      "three\n",
      "execs\n",
      "in\n",
      "india\n",
      "in\n",
      "recent\n",
      "weeks,\n",
      "according\n",
      "to\n",
      "bloomberg\n",
      "sources.\n",
      "the\n",
      "departures\n",
      "include\n",
      "the\n",
      "national\n",
      "sales\n",
      "and\n",
      "distribution\n",
      "chief,\n",
      "the\n",
      "commercial\n",
      "channels\n",
      "and\n",
      "midmarket\n",
      "business\n",
      "head,\n",
      "and\n",
      "the\n",
      "leader\n",
      "of\n",
      "telecom\n",
      "carrier\n",
      "sales.\n",
      "apple\n",
      "has\n",
      "struggled\n",
      "-\n",
      "to\n",
      "gain\n",
      "+\n",
      "traction\n",
      "+\n",
      "in\n",
      "india\n",
      "due\n",
      "to\n",
      "high\n",
      "import\n",
      "tariffs\n",
      "and\n",
      "a\n",
      "consumer\n",
      "base\n",
      "that\n",
      "prefers\n",
      "+\n",
      "cheaper\n",
      "+\n",
      "smartphones.\n",
      "apple\n",
      "has\n",
      "a\n",
      "singledigit\n",
      "market\n",
      "share\n",
      "in\n",
      "the\n",
      "country.\n",
      "sources\n",
      "say\n",
      "apple\n",
      "has\n",
      "been\n",
      "slow\n",
      "-\n",
      "to\n",
      "cultivate\n",
      "local\n",
      "business\n",
      "relationships\n",
      "and\n",
      "has\n",
      "had\n",
      "difficulties\n",
      "-\n",
      "understanding\n",
      "the\n",
      "market.\n",
      "michel\n",
      "coulomb\n",
      "took\n",
      "over\n",
      "as\n",
      "head\n",
      "of\n",
      "indian\n",
      "operations\n",
      "last\n",
      "december.\n",
      "coulomb\n",
      "has\n",
      "been\n",
      "with\n",
      "apple\n",
      "since\n",
      "2003\n",
      "and\n",
      "was\n",
      "previously\n",
      "the\n",
      "south\n",
      "asia\n",
      "managing\n",
      "director.\n",
      "apple\n",
      "shares\n",
      "are\n",
      "up\n",
      "0.2\n",
      "premarket\n",
      "to\n",
      "191.65.\n",
      "previously:\n",
      "apple\n",
      "roundup:\n",
      "production\n",
      "starts\n",
      "on\n",
      "iphone\n",
      "6s\n",
      "in\n",
      "india,\n",
      "barclays\n",
      "forecasts\n",
      "june\n",
      "q\n",
      "june\n",
      "26\n",
      "POSITIVE COUNT:4\n",
      "NEGATIVE COUNT:-6\n",
      "SUM IS:-2\n"
     ]
    }
   ],
   "source": [
    "# Read the tweet file and identify number of positive and negative words while taking care of stop words\n",
    "positive_tweets=0\n",
    "negative_tweets=0\n",
    "##defining clean(word) function for the purpose of removing special characters\n",
    "def clean(word):\n",
    "    symbols = ['!','@','#','$','%','_','^','&','*','(',')','-','+','=','?']\n",
    "    for symbol in symbols:\n",
    "        word = word.replace(symbol,'')\n",
    "    return word\n",
    "\n",
    "with open(\"C:\\\\Users\\\\shikh\\\\Desktop\\\\Practicum\\\\Apple_India.txt\",'r') as tweets:            ###opening original tweet file\n",
    "    for line in tweets:                         \n",
    "        line = line.strip()                       ##stripping characters before and after the line\n",
    "        words = line.split()                      ## getting words from  the lines\n",
    "        for word in words:\n",
    "            word = word.lower()                 ##getting lower case of the word\n",
    "            word = clean(word)                  ##applying clean function to the words\n",
    "            print(word)\n",
    "            if word not in stop_words:          ##check if word  is not in stop_words dictionary\n",
    "                if word in positive_words:      ## check if the word is in positive_words dictionary \n",
    "                    print('+')\n",
    "                    positive_tweets = positive_tweets +1       ###if yes add 1 to count of positive_tweets\n",
    "                elif word in negative_words:                    ##if no checck if the words is in negative_words dictionary\n",
    "                    negative_tweets = negative_tweets -1        ## if yes substract 1 from the count of negative_tweets\n",
    "                    print('-')\n",
    "                    \n",
    "##concatenating string POSITIVE COUNT:with string of positive_tweets\n",
    "print(\"POSITIVE COUNT:\" + str(positive_tweets))  \n",
    "##concatenating string NEGATIVE COUNT:with string of negative_tweets\n",
    "print(\"NEGATIVE COUNT:\" + str(negative_tweets))           \n",
    "\n",
    "# Concatening and printing sum of positive and negative tweets:\n",
    "\n",
    "print(\"SUM IS:\" + str(positive_tweets+negative_tweets))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "##Therefore, the sum of all values observed  is -2 \n",
    "##The general sentiment is negative.\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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}