Uncovering the community structure associated with the diffusion dynamics of networks

TL;DR

This paper reveals that the community structure of complex networks can be uncovered by analyzing diffusion dynamics, specifically through stable equilibrium states and conductance optimization, outperforming traditional methods.

Contribution

It introduces a novel approach linking diffusion processes to community detection via conductance optimization, improving accuracy over modularity-based methods.

Findings

Conductance optimization outperforms modularity in community detection

Stable local equilibrium states reveal intrinsic community structures

Method effective on both benchmark and real-world networks

Abstract

As two main focuses of the study of complex networks, the community structure and the dynamics on networks have both attracted much attention in various scientific fields. However, it is still an open question how the community structure is associated with the dynamics on complex networks. In this paper, through investigating the diffusion process taking place on networks, we demonstrate that the intrinsic community structure of networks can be revealed by the stable local equilibrium states of the diffusion process. Furthermore, we show that such community structure can be directly identified through the optimization of the conductance of network, which measures how easily the diffusion occurs among different communities. Tests on benchmark networks indicate that the conductance optimization method significantly outperforms the modularity optimization methods at identifying the community structure of networks. Applications on real world networks also demonstrate the effectiveness of the conductance optimization method. This work provides insights into the multiple topological scales of complex networks, and the obtained community structure can naturally reflect the diffusion capability of the underlying network.