Copula Mixed-Membership Stochastic Blockmodel for Intra-Subgroup Correlations

TL;DR

This paper introduces a novel extension of the MMSB model that incorporates copula functions to explicitly model correlations in subgroup memberships within social networks, improving link prediction accuracy.

Contribution

It proposes the copula-based MMSB (cMMSB) model that captures intra-subgroup correlations, addressing a limitation of the traditional MMSB model.

Findings

cMMSB outperforms existing models in link prediction tasks

The model effectively captures subgroup correlations in synthetic and real data

Flexible use of different copula functions enhances modeling capabilities

Abstract

The \emph{Mixed-Membership Stochastic Blockmodel (MMSB)} is a popular framework for modeling social network relationships. It can fully exploit each individual node's participation (or membership) in a social structure. Despite its powerful representations, this model makes an assumption that the distributions of relational membership indicators between two nodes are independent. Under many social network settings, however, it is possible that certain known subgroups of people may have high or low correlations in terms of their membership categories towards each other, and such prior information should be incorporated into the model. To this end, we introduce a \emph{Copula Mixed-Membership Stochastic Blockmodel (cMMSB)} where an individual Copula function is employed to jointly model the membership pairs of those nodes within the subgroup of interest. The model enables the use of various Copula functions to suit the scenario, while maintaining the membership's marginal distribution, as needed, for modeling membership indicators with other nodes outside of the subgroup of interest. We describe the proposed model and its inference algorithm in detail for both the finite and infinite cases. In the experiment section, we compare our algorithms with other popular models in terms of link prediction, using both synthetic and real world data.