Inferring Social Relationships across Social Networks for Viral Marketing

Abstract

Node classification in social networks is a useful and important technique that has been widely studied in recent years. Many existing node classification methods mainly focus on exploiting structural and attribute information to identify node classes. However, the information in an emerging information network is usually limited. For example, a social networking platform that includes a few registered users(referred to as active users) and a significant amount of new comers (referred to as non-active users) with very sparse interactions among registered users. Under this circumstances, distinguishing the users that is likely to be active in the future from large-scale new comers becomes challenging. In this paper, we propose a hybrid classification model, which can distinguish whether a nonactive user will become an active user in the future by incorporating multiple relations through a unified ranking measure. Particularly, given a friendship network and a mobile communication network, we aim to discover a small set of users, who are likely to become active users in the future, from a massive amount of non-active users. X We conducted extensive experiments to demonstrate the effectiveness of our hybrid ranking model as well as report several empirical observations from real data sets.