Memetic algorithm for community detection in networks

Abstract

Community structure is one of the most important properties in networks, and community detection has received an enormous amount of attention in recent years. Modularity is by far the most used and best known quality function for measuring the quality of a partition of a network, and many community detection algorithms are developed to optimize it. However, there is a resolution limit problem in modularity optimization methods. In this study, a memetic algorithm, named Meme-Net, is proposed to optimize another quality function, modularity density, which includes a tunable parameter that allows one to explore the network at different resolutions. Our proposed algorithm is a synergy of a genetic algorithm with a hill-climbing strategy as the local search procedure. Experiments on computer-generated and real-world networks show the effectiveness and the multiresolution ability of the proposed method.

Community Detection in Complex Networks based on Improved Genetic Algorithm and Local Optimization

“…On the basis of references [9][10][11][12][13][14][15] and analysis of them, this paper gives the parameter settings of IGALO algorithm and the parameter settings are presented in Table 2. Mutation rate 30%…”

“…13.end for End Thus, in the process of promoting the mutation of node label, mutation strategy of node local optimization not only considers the similarity of nodes and intra-community connections, but also considers the connection density of nodes and community. Here Figure 2 is the example to be illustrated.…”

“…And it also proposes the strategy of multi-individual clustering combination crossover with the combination of function Q and completes the task of community detection with the combination of local mutation strategy. Meme-Net algorithm [13] is proposed by Gong et. al., In order to obtain better community structure, hill climbing algorithm is used to further process the elite individuals obtained through genetic algorithm.…”

A Novel Fuzzy Clustering Recommendation Algorithm Based On Pso

“…is the objective function; the objective function F defines a mapping function [8] and m targets which need to be optimized;…”

“…Compared with single objective optimization [8], the complexity of multiobjective optimization has greatly increased. It needs to optimize multiple objectives, which are not comparable, and even conflicting.…”