API_Twitter

SENTIMENT ANALYSIS With Python Using TwitterAPI

About our Project

Sentiment analysis is the use of Natural Language Processing, text extraction to study effective state of subjective information, in this case tweets from TwitterAPI which provides endpoints for accessing and pulling tweets from twitter. We also had an option of using Stocktwits, a platform containing information for traders and communications on trends. Our group focused mainly with TwitterAPI. Pre-requisites: • Python: preferably python 3 • Twitter developers authenticated account • TextBlob package • Tweepy • Basic knowledge on Python

To get access to tweets from Twitter, we needed to create a developers account on Twitter. It provides ability to create an application which authenticates; provide keys to authenticate us to access tweets. Python libraries used are tweepy and TextBlob Below is our code: We tried as much to comment on the code from textblob import TextBlob import tweepy import sys

#set up our keys from twitter developers account api_key = "1INSA5zlzR1mBpEwwaGNICHPI" api_key_secret = "tnbv7zw0LvLL2qiV1IHdztQIsdm80Z3HI6vmuLOomPZTJBrrte" access_token = "1175064453743828992-hoTX48Iy5vsjsUqu6wYbTiJQ4qjJfD" access_token_secret = "dytyZ2wdrGYuqbhWxpTLI6mkvFyptoAYhCRxBM41S2Bpv"

#create an authentication handler auth_handler = tweepy.OAuthHandler(consumer_key = api_key, consumer_secret= api_key_secret)

auth_handler.set_access_token(access_token, access_token_secret) api = tweepy.API(auth_handler)

#define our search term and amount of tweets we would like to work with search_term = 'sad' tweet_amount = 10

#we will use Cursor, a functionality of out package tweepy to initiate our search for tweets tweets = tweepy.Cursor(api.search, q=search_term, lang= 'en').items(tweet_amount)

#we also need to set the variables below to make sense of output from our tweets polarity = 0

positive = 0 neutral = 0 negative = 0

for tweet in tweets: #print(tweet.text) #this gives us tweets including hashtags and # irrelevant content we do not need and so we need to clean tweets #we receive from twitter final_text = tweet.text.replace('RT', '') if final_text.startswith(' @'): position = final_text.index(':') final_text = final_text[position+2:] #print(final_text) #prints our cleaned texts

    #We are ready to do sentiment analysis of our tweets
    analysis = TextBlob(final_text)
    tweet_polarity = analysis.polarity
    if tweet_polarity > 0:
        positive +=1
    elif tweet_polarity < 0:
        negative +=1
    else:
        neutral +=1
    polarity += tweet_polarity
    print(final_text)

print(polarity) print(f'amount of positive tweets: {positive}') print(f'amount of negative tweets: {negative}') print(f'amount of neutral tweets: {neutral}')