Community Detection Algorithm Based On Complex Network

The characteristics of complex network structure and the causes of some network phenomena can be revealed through the study of community discovery in complex networks,and finding its implicit model and predict its behavior have wide application value in social network,biomolecule network and Internet.However,due to the complex and complex structure of the network,the universality of the algorithm and the accuracy of the results are reduced.The dynamic changes of the network will destroy the original community structure and re-divide the community structure will produce a lot of computational overhead.Therefore,to ensure the accuracy of the algorithm while improving the efficiency of the field is an urgent solution to the technical problems.Based on the research status of community discovery algorithm,an adaptive community discovery algorithm KDED is proposed,which is based on the idea of density peak clustering,guarantees the high efficiency and increases the adaptability with the thermal diffusion model.Meanwhile,a dynamic community discovery algorithm D-KDED based on local increment is proposed,based on the assumption of short-term smoothness.The incremental algorithm is used to divide the community and reduce the time cost.The simulation results show that the KDED algorithm proposed in this paper has a very high degree of modularity index on the real data set and the artificial data set.In addition,the fuzzy network with relatively weak community structure has excellent community structure,The D-KDED algorithm proposed in this paper has a good modularity index and the highest computational efficiency in the real dynamic network.In the large-scale network with sparse network structure,the D-KDED algorithm has the highest stability and accuracy,which can effectively discover the community structure,while having the lowest time cost.