|
Community detection in networks is to find groups of nodes with similar characteristics, which is commonly defined as finding dense connection groups in undirected networks. However, communities in directed networks usually represent group action patterns because of asymmetric relations, which is difficult to capture using traditional algorithms. In this paper, a Gamma-Poisson blockmodel is proposed for community detection in directed networks, which can model not only assortative communities but also communities with various connectivity patterns due to a block matrix. The model can also be extended to undirected networks if we set the block matrix symmetric, and for assortative community detection task if we set the block matrix diagonal. We develop an efficient Gibbs sampling algorithm for the inference work, which can scale to large sparse networks since only links are considered during each iteration. We compare our model with several previous methods and results demonstrate our advantages on a variety of real-world networks. |
|
Keywords:Artificial Intelligence; Community detection; Directed networks; Gibbs sampling |
|