Abstract
The problem of community detection in networks is usually formulated as finding a single partition of the network into some “correct” number of communities. We argue that it is more interpretable and in some regimes more accurate to construct a hierarchical tree of communities instead. This can be done with a simple top-down recursive partitioning algorithm, starting with a single community and separating the nodes into two communities by spectral clustering repeatedly, until a stopping rule suggests there are no further communities. This class of algorithms is model-free, computationally efficient, and requires no tuning other than selecting a stopping rule. We show that there are regimes where this approach outperforms K-way spectral clustering, and propose a natural framework for analyzing the algorithm’s theoretical performance, the binary tree stochastic block model. Under this model, we prove that the algorithm correctly recovers the entire community tree under relatively mild assumptions. We apply the algorithm to a gene network based on gene co-occurrence in 1580 research papers on anemia, and identify six clusters of genes in a meaningful hierarchy. We also illustrate the algorithm on a dataset of statistics papers. Supplementary materials for this article are available online.
Supplementary Materials
The online supplementary materials contain our proofs for the theoretical results, additional details of the Anemia example, and one additional example of using HCD to analyze a citation network.
Acknowledgments
We thank the associate editor and the referees for their helpful and constructive comments.
Funding
T. Li was supported in part by an NSF grant (DMS-2015298) and the Quantitative Collaborative Award from the College of Arts and Sciences at the University of Virginia. K. Van den Berge is a postdoctoral fellow of the Belgian American Educational Foundation (BAEF) and is supported by the Research Foundation Flanders (FWO), grant 1246220N. P. Sarkar was supported in part by an NSF grant (DMS-1713082). P. Bickel is supported in part by an NSF grant (DMS-1713083). E. Levina is supported in part by NSF grants (DMS-1521551 and DMS-1916222) and an ONR grant (N000141612910).