Who is suitable to be followed back when you are a Twitter interested in Politics?

Abstract

The degree of citizens participation and political involvement within democratic societies may vary from one person to another. Similarly, the diversity of political engagement of users might be observed in online platforms. For example, Twitter users might be characterized as highly politically active or poorly politically active citizens, according to their Twitter activity. Identifying the level of interest in politics of a user would be relevant to provide them with meaningful recommendations such as political actors to follow, tweets talking about politics, and political-oriented lists, among others. However, due to the information overload and the wide range of topics posted on Twitter, generating personalized political-related suggestions becomes a problem. In this paper, we address this challenge by presenting a preliminary work where we i) identify the degree of interest in politics (DoIP) of a sample of Twitter users and ii) measure the correlation of this degree with their Twitter friends' DoIP, aiming to use it in following back recommendations. Both i) and ii) can be considered as core bases when designing a following back recommender system for users interested in politics. This research approach is novel with respect to the state of the art given the current lack of studies in following back recommendations for Twitter users. Experiments on real data in the context of politics in Ecuador show the effectiveness of our approach in identifying the DoIP of Twitter citizens and the positive association of it with their friends'.