Sentiment based Analysis of Tweets during the US Presidential Elections

Abstract

In a relatively short period of time, social media has gained significant importance as a mass communication and public engagement tool for political and governance purposes. Rapid dissemination of information through social media platforms such as Twitter, provides politicians and campaigners with the ability to broadcast their message to a wide audience instantly and directly while bypassing the traditional media channels. In this paper, we investigate the nature and characteristics of the political discourse that took place on Twitter during the American Presidential elections of November 2016. The goal of this study is to perform exploratory sentiment based analysis of Twitter data that was gathered both before and after the Election Day. Our objective is to identify the nature and sentiment of discussions along with understanding the behavior of users with respect to their Twitter profile and associated attributes of their tweets. We also aim to inspect popular Twitter discussion topics and their relation with important news and events occurring simultaneously.