Determining the Number of Communities in Degree-corrected Stochastic Block Models

TL;DR

This paper introduces a new method combining spectral clustering and binary segmentation to accurately estimate the number of communities in degree-corrected stochastic block models, with proven consistency and good performance in semi-dense networks.

Contribution

The paper presents a novel approach for estimating community numbers using a pseudo likelihood ratio statistic with theoretical guarantees and practical validation.

Findings

Method guarantees an upper bound for over-fitted models

Estimator is consistent under mild degree growth conditions

Performs well in simulations and real-world network analysis

Abstract

We propose to estimate the number of communities in degree-corrected stochastic block models based on a pseudo likelihood ratio statistic. To this end, we introduce a method that combines spectral clustering with binary segmentation. This approach guarantees an upper bound for the pseudo likelihood ratio statistic when the model is over-fitted. We also derive its limiting distribution when the model is under-fitted. Based on these properties, we establish the consistency of our estimator for the true number of communities. Developing these theoretical properties require a mild condition on the average degrees -- growing at a rate no slower than log(n), where n is the number of nodes. Our proposed method is further illustrated by simulation studies and analysis of real-world networks. The numerical results show that our approach has satisfactory performance when the network is semi-dense.