Detecting network communities by propagating labels under constraints

TL;DR

This paper reformulates the label propagation algorithm for community detection as an optimization problem, identifies its drawbacks, and proposes constrained variants that improve community quality, including modularity maximization, for bipartite and unipartite networks.

Contribution

It introduces a new constrained optimization framework for label propagation, addressing limitations of existing methods and enhancing community detection quality.

Findings

Modified algorithms better optimize community quality measures.

Proposed methods perform well on bipartite and unipartite networks.

Addressed the disparity between objective function value and community quality.

Abstract

We investigate the recently proposed label-propagation algorithm (LPA) for identifying network communities. We reformulate the LPA as an equivalent optimization problem, giving an objective function whose maxima correspond to community solutions. By considering properties of the objective function, we identify conceptual and practical drawbacks of the label propagation approach, most importantly the disparity between increasing the value of the objective function and improving the quality of communities found. To address the drawbacks, we modify the objective function in the optimization problem, producing a variety of algorithms that propagate labels subject to constraints; of particular interest is a variant that maximizes the modularity measure of community quality. Performance properties and implementation details of the proposed algorithms are discussed. Bipartite as well as unipartite networks are considered.