LabelRank: A Stabilized Label Propagation Algorithm for Community Detection in Networks

TL;DR

LabelRank is a stabilized label propagation algorithm that reliably detects communities in large-scale networks, improving upon traditional methods by controlling propagation dynamics and ensuring consistent results.

Contribution

The paper introduces LabelRank, a novel algorithm with operators that stabilize label propagation, addressing randomness and improving community detection quality.

Findings

Significantly improves community detection accuracy

Produces consistent community structures across runs

Outperforms traditional label propagation algorithms

Abstract

An important challenge in big data analysis nowadays is detection of cohesive groups in large-scale networks, including social networks, genetic networks, communication networks and so. In this paper, we propose LabelRank, an efficient algorithm detecting communities through label propagation. A set of operators is introduced to control and stabilize the propagation dynamics. These operations resolve the randomness issue in traditional label propagation algorithms (LPA), stabilizing the discovered communities in all runs of the same network. Tests on real-world networks demonstrate that LabelRank significantly improves the quality of detected communities compared to LPA, as well as other popular algorithms.