Fusion of Heterogeneous Social Networks for Synergistic Knowledge Discovery

In this thesis, we will focus on introducing the information fusion learning works done based on online social media data. To enjoy more social network services, people are usually involved in multiple online social networks simultaneously, such as Facebook, Twitter and Foursquare. Our work in this thesis covers five strongly correlated research directions in the study of information networks fusion and mining, including network alignment, link prediction, community detection, information diffusion, and network embedding. These application tasks are fundamental problems in social network studies, which together with the network alignment problem will form the backbone of the multiple social network fusion learning ecosystem.