Complex Network Community Detection Algorithm Resolution And Limit Research

The problem of complex network community detection has been a hot topic in the field of network research.For example,community detection algorithms based on modularity and Infomap community detection algorithms based on information theory have achieved good results in application.However,due to the nature of modularity function,the algorithm based on modularity optimization may not be able to recognize the small communities in complex networks.A similar resolution limitation problem exists in the Infomap algorithm.As for the resolution limit of community detection algorithm,this paper presents the resolution limit of community recognition based on modularity optimization.Similarly,we further estimated the resolution limit of Infomap algorithm based on information coding theory.By comparing the resolution limits of the two algorithms,we find that Infomap algorithm is better than modularity algorithm in detecting the network community structure.Finally,through the simulation of artificial network data and the empirical analysis of real network data,it further confirms the intrinsic superiority of Infomap algorithm proposed in this paper in terms of the resolution limit.