Inferring political preferences of active content consumers in Twitter

Abstract

The growth of user engagement in online social networks has generated a tremendous amount of content regarding various topics. This rich content helps businesses to infer interesting information about public opinions and preferences of OSN users to serve their customers with customized services. Also, this inferred information can be used for different prediction purposes, such as predicting the possible outcome of an election. Despite the huge increase in the amount of produced content in OSNs, many users tend to consume content on certain topics rather than provide content themselves. Therefore, it is a challenge to discover preferences of content consumers who are silent on a given topic. In this thesis, a novel approach is proposed that predicts personal preferences of content consumers through what they read rather than what they write. In other words, in this study it is shown that only relying on followees to predict preferences of content consumers leads to promising results.