Analyzing political sentiment using Twitter data

Abstract

Today microblogging has become a very common platform for exchanging opinion among us. Many users exchange their thoughts on various aspects of their activity. Consequently, microblogging Web sites are the substantial origin of information for sentiment analysis and opinion mining. Twitter is a famous microblogging Web site where 500 million tweets are posted every day. In this manuscript, we summarize the data set of Twitter messages related to recent 14th Gujarat Legislative Assembly Election, 2017, for predicting the chances of winning party by utilizing public’s opinion. We use NRC Emotion Lexicon to determine the overall tone of the event by eight emotions. Furthermore, we use a deep learning tool named ParallelDots AI APIs by ParallelDots, Inc. that can analyze the sentiment into positive, negative, and neutral. This tool helped to extract various people’s sentiment and summarize the results for further decision making.