Abstract
One of the most fundamental problems in network study is community detection. The stochastic block model (SBM) is a widely used model, and various estimation methods have been developed with their community detection consistency results unveiled. However, the SBM is restricted by the strong assumption that all nodes in the same community are stochastically equivalent, which may not be suitable for practical applications. We introduce a pairwise covariates-adjusted stochastic block model (PCABM), a generalization of SBM that incorporates pairwise covariate information. We study the maximum likelihood estimators of the coefficients for the covariates as well as the community assignments, and show they are consistent under suitable sparsity conditions. Spectral clustering with adjustment (SCWA) is introduced to efficiently solve PCABM. Under certain conditions, we derive the error bound of community detection for SCWA and show that it is community detection consistent. In addition, we investigate model selection in terms of the number of communities and feature selection for the pairwise covariates, and propose two corresponding algorithms. PCABM compares favorably with the SBM or degree-corrected stochastic block model (DCBM) under a wide range of simulated and real networks when covariate information is accessible. Supplementary materials for this article are available online.
Supplementary Materials
The supplementary material contains the proofs of the theoretical results, presents some technical lemmas, and shows additional simulation results.
Disclosure Statement
The authors report there are no competing interests to declare.
Acknowledgments
We thank the editor, the AE, and anonymous reviewers for their insightful comments which have greatly improved the scope and quality of the article.
Notes
1 Note that these are “edge-level” covariates instead of the nodal or vertex-level covariates that are often considered in other parts of the literature. Having said that, one can incorporate nodal information into our model by converting it into pairwise covariates, where an example will be presented in Section 8.1.