A survey on political event analysis in Twitter

Abstract

This short survey paper attempts to provide an overview of the most recent research works on the popular politics domain within the framework of the Twitter social network. Given both the political turmoil that arouse at the end of 2016 and early 2017, and the increasing popularity of social networks in general, and Twitter, in particular, we feel that this topic forms an attractive candidate for fellow data mining researchers that came into sight over the last few months. Herein, we start by presenting a brief overview of our motivation and continue with basic information on the Twitter platform, which constitutes two clearly identifiable components, namely as an online news source and as one of the most popular social networking sites. Focus is then given to research works dealing with sentiment analysis in political topics and opinion polls, whereas we continue by reviewing the Twittersphere from the computational social science point of view, by including behavior analysis, social interaction and social influence identification methods and by discerning and discriminating its useful types within the social network, thus envisioning possible further utilization scenarios for the collected information. A short discussion on the identified conclusions and a couple of future research directions concludes the survey.