Kristina Lerman

A Visibility-based Model for Link Prediction in Social Media

TitleA Visibility-based Model for Link Prediction in Social Media
Publication TypeConference Paper
Year of Publication2014
AuthorsL. Zhu, and K. Lerman
Conference NameProceedings of the ASE/IEEE Conference on Social Computing

A core task of social network analysis is to predict the formation of new social links. In the context of social media, link prediction serves as the foundation for forecasting the evolution of the follower graph and predicting interactions and the flow of information between users. Previous link prediction methods have generally represented the social network as a graph and leveraged topological and semantic measures of similarity between two nodes to evaluate the probability of link formation. In this work, we suggest another link creation mechanism for social media wherein a user v creates a link to user u after seeing u�s name on his or her screen. In other words, visibility of a user (name) is a necessary condition for new link formation. We propose a visibility-based model for link prediction, which estimates the probability of a user views another user�s name, and use this model to predict new links. We further estimate a set of parameters in the proposed visibility-based model by a Maximum-Likelihood approach with a MM algorithm. Empirical study shows that the proposed model can more accurately predict both follow and co-mention links than alternative state-of-the-art methods. Our work suggests that the effort required to discover a new social contact is negatively correlated with link formation, and the easier it is to discover a user, the higher the likelihood a link to the user will be created.