Model And Algorithm Of Community Division Based On Weak Community Definition

Various networks in real life,such as social networks,transportation networks,biological networks can be abstracted as a theoretical tool of complex networks for research.The community structure is one of the important characteristics of complex networks.It reduces the complicated network to seemingly relatively independent but interconnected communities.The research on the division of complex network community structure has important theoretical significance and practical value.For the complex network community structure,different scholars have given corresponding definitions based on different evaluation criteria.When studying the network community structure,there may be a phenomenon of incorrect division of nodes.Based on this,this paper constructs the framework of the weak community definition integer programming model,and then analyzes the community structure of the network based on the theory of spectral average method.The integer programming model constructed in the article is a discrete problem,and it is difficult to solve a network with a large number of nodes.This model is transformed into a continuous optimization approximate model,the corresponding block coordinate descent algorithm solution model is given,and the relationship between the original problem and the approximate problem solution is analyzed.In this paper,artificial network and real-world networks are selected to verify the feasibility of the community division model and algorithm based on the definition of weak communities.In the real-world networks,the results of the network community division between the method in this paper and the spectral average method are compared.The experimental results show that the results obtained by the method proposed in this paper meet the definition of weak community and are qualified community structures.From the results of the division,its accuracy is better than that of the spectral average method.