Privacy Leakages about Political Beliefs through Analysis of Twitter Followers

Briola, Helen ; Drosatos, George ; Stamatelatos, Giorgos ; Gyftopoulos, Sotirios ; Efraimidis, Pavlos S. (2018-11)


In this paper, we focus on privacy leakages about Twitter usersand show that simply establishing follower and friend connectionsin the Twitter network might be enough to reveal sensitive in-formation about the political beliefs of a user. More precisely, wecreate a Twitter dataset containing the twitter nodes of Greek newsmedia and politicians, as well as their followers and present twoapproaches to predict the political beliefs of a Twitter user and thepoliticians the user will potentially follow. The intuition behindboth approaches is the saying“Show me who your friends are, andI will tell you who you are”. In the first approach, the politiciansfollowed by the followers of the user are shown to be correlated tothe politicians that the user follows. In the second one, the mediathat the user follows are used to predict the politicians that the userfollows. The experimental results of both approaches confirm ourclaims and show that significant privacy leakages are possible.

Except where otherwise noted, this item's license is described as