Dynamic Network Sampling for Community Detection

TL;DR

This paper introduces a dynamic sampling method for community detection in large networks, optimizing resource use while maintaining accurate block recovery in stochastic blockmodels.

Contribution

It proposes a Chernoff-optimal dynamic sampling scheme justified theoretically and validated on real datasets, improving efficiency in community detection tasks.

Findings

The method effectively identifies influential vertices for block structure.

It reduces observation costs while preserving block recovery accuracy.

Theoretical analysis confirms optimality via Chernoff information.

Abstract

We propose a dynamic network sampling scheme to optimize block recovery for stochastic blockmodel (SBM) in the case where it is prohibitively expensive to observe the entire graph. Theoretically, we provide justification of our proposed Chernoff-optimal dynamic sampling scheme via the Chernoff information. Practically, we evaluate the performance, in terms of block recovery, of our method on several real datasets from different domains. Both theoretically and practically results suggest that our method can identify vertices that have the most impact on block structure so that one can only check whether there are edges between them to save significant resources but still recover the block structure.