Skip to content

Instantly share code, notes, and snippets.

@Omkaragrawal
Created September 29, 2018 05:25
Show Gist options
  • Save Omkaragrawal/7afce90d23ced06945a46759aae4c1df to your computer and use it in GitHub Desktop.
Save Omkaragrawal/7afce90d23ced06945a46759aae4c1df to your computer and use it in GitHub Desktop.
Sentimental analysis twitter
import re
import tweepy
from tweepy import OAuthHandler
from textblob import TextBlob
 
class TwitterClient(object):
    '''
   Generic Twitter Class for sentiment analysis.
   '''
    def __init__(self):
        '''
       Class constructor or initialization method.
       '''
        # keys and tokens from the Twitter Dev Console
        consumer_key = 'XXXXXXXXXXXXXXXXXXXXXXXX'
        consumer_secret = 'XXXXXXXXXXXXXXXXXXXXXXXXXXXX'
        access_token = 'XXXXXXXXXXXXXXXXXXXXXXXXXXXX'
        access_token_secret = 'XXXXXXXXXXXXXXXXXXXXXXXXX'
 
        # attempt authentication
        try:
            # create OAuthHandler object
            self.auth = OAuthHandler(consumer_key,consumer_secret)
            # set access token and secret
            self.auth.set_access_token(access_token,access_token_secret)
            # create tweepy API object to fetch tweets
            self.api = tweepy.API(self.auth)
        except:
            print("Error: Authentication Failed")
 
    def clean_tweet(self, tweet):
        '''
       Utility function to clean tweet text by removing links, special characters
       using simple regex statements.
       '''
        return ' '.join(re.sub("(@[A-Za-z0-9]+)|([^0-9A-Za-z \t])
                                   |(\w+:\/\/\S+)", " ",tweet).split())
 
    def get_tweet_sentiment(self, tweet):
        '''
       Utility function to classify sentiment of passed tweet
       using textblob's sentiment method
       '''
        # create TextBlob object of passed tweet text
        analysis = TextBlob(self.clean_tweet(tweet))
        # set sentiment
        if analysis.sentiment.polarity > 0:
            return 'positive'
        elif analysis.sentiment.polarity == 0:
            return 'neutral'
        else:
            return 'negative'
 
    def get_tweets(self, query, count = 10):
        '''
       Main function to fetch tweets and parse them.
       '''
        # empty list to store parsed tweets
        tweets = []
 
        try:
            # call twitter api to fetch tweets
            fetched_tweets = self.api.search(q = query,count = count)
 
            # parsing tweets one by one
            for tweet in fetched_tweets:
                # empty dictionary to store required params of a tweet
                parsed_tweet = {}
 
                # saving text of tweet
                parsed_tweet['text'] = tweet.text
                # saving sentiment of tweet
                parsed_tweet['sentiment'] =self.get_tweet_sentiment(tweet.text)
 
                # appending parsed tweet to tweets list
                if tweet.retweet_count > 0:
                    # if tweet has retweets, ensure that it is appended only once
                    if parsed_tweet not in tweets:
                        tweets.append(parsed_tweet)
                else:
                    tweets.append(parsed_tweet)
 
            # return parsed tweets
            return tweets
 
        except tweepy.TweepError as e:
            # print error (if any)
            print("Error : " + str(e))
 
def main():
    # creating object of TwitterClient Class
    api = TwitterClient()
    # calling function to get tweets
    tweets = api.get_tweets(query = 'Donald Trump', count =200)
 
    # picking positive tweets from tweets
    ptweets = [tweet for tweet in tweets iftweet['sentiment'] == 'positive']
    # percentage of positive tweets
    print("Positive tweets percentage: {} %".format(100*len(ptweets)/len(tweets)))
    # picking negative tweets from tweets
    ntweets = [tweet for tweet in tweets iftweet['sentiment'] == 'negative']
    # percentage of negative tweets
    print("Negative tweets percentage: {} %".format(100*len(ntweets)/len(tweets)))
    # percentage of neutral tweets
    print("Neutral tweets percentage: {} % \
       ".format(100*len(tweets - ntweets - ptweets)/len(tweets)))
 
    # printing first 5 positive tweets
    print("\n\nPositive tweets:")
    for tweet in ptweets[:10]:
        print(tweet['text'])
 
    # printing first 5 negative tweets
    print("\n\nNegative tweets:")
    for tweet in ntweets[:10]:
        print(tweet['text'])
 
if __name__ == "__main__":
    # calling main function
    main()
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment