Inferring Social Relations from Online and Communication Networks

Abstract

In this work analyzed the interplay between social relations in the form of friendship ties, attributes and interaction in online social networks. In this context we analyzed composition of social circles in online social networks and showed that social circles are homophilious with respect to at least one node attribute. We showed that using the right combination of network and interaction features, links can be inferred in online covert networks. We also analyzed longitudinal dyadic interaction on cellular networks and proposed a call prediction model.