Probabilistic Community and Role Model for Social Networks

Yu Han, Jie Tang
2015 Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD '15  
Numerous models have been proposed for modeling social networks to explore their structure or to address application problems, such as community detection and behavior prediction. However, the results are still far from satisfactory. One of the biggest challenges is how to capture all the information of a social network such as links, communities, user attributes, roles and behaviors, in a unified manner. In this paper, we propose a unified probabilistic framework, the Community Role Model
more » ... , to model a social network. CRM incorporates all the information of nodes and edges that form a social network. We propose methods based on Gibbs sampling and an EM algorithm to estimate model parameters and fit our model to real social networks. Real data experiments show that CRM can be used not only to represent a social network, but also to handle various application problems with better performance than a baseline model, without any modification to the model.
doi:10.1145/2783258.2783274 dblp:conf/kdd/HanT15 fatcat:wvqpsfzmgva35mdydjv424zvhu