|
Abstract:
|
The increasing prevalence of network data in a vast variety of fields and the need to extract useful information out of them have spurred fast developments in models and algorithms for the inference of networks. Among the specific learning tasks with network data, community detection which divides nodes into clusters or ‘communities’, has arguably received the most attention in the scientific community. In addition to the observed network, their is often additional covariates information available which should ideally be incorporated when performing community detection. We add to a limited literature on community detection for networks with covariates by proposing a Bayesian community detection in which the effects of the covariates are incorporated via a covariates dependent random partition prior, under which a block model is assumed. One of the distinctive features of our models is that it has the ability to learn the number of the community via posterior inference without having to assume that it is known. Our extensive simulation demonstrated our superior performance over existing methods. The developed MCMC algorithm is very efficient with very good mixing properties.
|
