On New Group Popularity Prediction in Event-Based Social Networks

Abstract

Event-based social networks (EBSN) have recently emerged as an important complement to online social networks. They enjoy the advantages of both online social networks and offline social communities: offline social events can be conveniently organized online, and users interact with each other face-to-face in the organized offline events. Although previous work has shown that member and structural features are important to the future popularity of groups in the EBSN, it is not yet clear how different member roles and the interplay between them contribute to group popularity. In this paper, we study a real-world dataset from Meetup—a popular EBSN platform—and propose a deep-neural-network-based method to predict the popularity of new Meetup groups. Our method uses group-level features specific to EBSNs, such as time and location of events in a group, as well as the structural features internal to a group, such as the inferred member roles in a group and social substructures among members. Empirically, our approach reduces the normalized root-mean-squared error of the popularity prediction (measured in RSVPs) of a group's future events by up to 12%, against the state-of-the-art baselines. Through case studies, our method also identifies member and structure patterns that are most predictive of a group's future popularity. Our study provides new understanding about what makes a group successful in the EBSN.