Identification of network modules by optimization of ratio association

Abstract

We introduce a novel method for identifying the modular structures of a network based on the maximization of an objective function: the ratio association. This cost function arises when the communities detection problem is described in the probabilistic autoencoder frame. An analogy with kernel k-means methods allows to develop an efficient optimization algorithm, based on the deterministic annealing scheme. The performance of the proposed method is shown on a real data set and on simulated networks.

References in corpus (9)

• Community structure in social and biological networks
• Modularity and community structure in networks
• Fast algorithm for detecting community structure in networks
• Cooperative Game Theory Approaches for Network Partitioning
• Functional cartography of complex metabolic networks
• Community detection in complex networks using Extremal Optimization
• Synchronization reveals topological scales in complex networks
• Detecting fuzzy community structures in complex networks with a Potts model
• Cost functions for pairwise data clustering

Cited by in corpus (4)

• Modularity and community detection in bipartite networks
• Kernel Granger causality and the analysis of dynamical networks
• An ensemble based on a bi-objective evolutionary spectral algorithm for graph clustering
• Detecting hierarchical and overlapping network communities using locally optimal modularity changes