Determining the popularity of political parties using twitter sentiment analysis

With the advancement in the Internet Technology, many people have started connecting to social networking websites and are using these microblogging websites to publically share their views on various issues such as politics, celebrity, or services like e-commerce. Twitter is one of those very popular microblogging website having 328 million of users around the world who posts 500 million of tweets per day to share their views. These tweets are rich source of opinionated User-Generated Content (UGC) that can be used for effective studies and can produce beneficial results. In this research, we have done Sentiment Analysis (SA) or Opinion Mining (OM) on user-generated tweets to get the reviews about major political parties and then used three algorithms, Support Vector Machine (SVM), Naïve Bayes Classifier, and k-Nearest Neighbor (k-NN), to determine the polarity of the tweet as positive, neutral, or negative, and finally based on these polarities we made a prediction of which party is likely to perform more better in the upcoming election.