A0370
Title: Two-way node popularity model for directed and bipartite networks
Authors: Ting Li - Hong Kong Polytechnic University (Hong Kong) [presenting]
Abstract: Community detection for directed and bipartite networks has raised lots of interest in recent decades. While many of the existing results introduced block-wise structure, most of them have the restriction that nodes in the same community behave identically or change uniformly in all communities. However, the heterogeneous node popularity is widely observed both in undirected and directed networks and has been studied under undirected scenarios. Motivated by the variability of node popularity in empirical directed networks, a novel probabilistic framework is proposed for directed network community detection, called the two-way node popularity model (TNPM). To fit the proposed model, the Rank One Approximation Algorithm (ROA) and establish the consistency of ROA is developed for community detection. In addition, an alternative computationally efficient algorithm, called Two-Stage Divided Cosine Algorithm (TSDC), is proposed to fit large-scale networks. Extensive numerical studies demonstrate the advantages of the proposed method in terms of both estimation accuracy and computation efficiency. The method is also applied to the Worldwide Trading Networks and MovieLens 100K Dataset, yielding some interesting findings.
