Complex Network Clustering Based On Multi-objective Optimization Algorithm

With the development of the Internet,the clustering problem of complex networks has become one of the research hotspots of scholars.Clustering complex networks helps to discover the closeness of the nodes and connected edges contained in different categories within the network,and then mines the feature similarity of such nodes;the nodes between different categories are different,and significant mining based on the differences is significant Sexual characteristics.This paper improves the multi-objective evolutionary algorithm,and derives two improved algorithms based on the multi-objective evolutionary framework,and applies them to complex networks and real artificial data sets,respectively.In order to accurately find the complex community structure,an improved multi-objective evolution complex network community detection algorithm is proposed.By generating multiple p parameters(biased parameters)at equal intervals in a certain range,they are substituted into the nearest neighbor propagation,AP)clustering algorithm.The semi-supervised clustering method is used to determine the number of clusters and generate initial populations,which overcomes the shortcomings of the traditional initial random clustering method which is unstable.The simulated annealing(Simulated Annealing,SA)algorithm is used to improve the multi-objective evolutionary algorithm to improve the population search ability and prevent the optimization process from falling into the local optimal solution.Compared with the experimental results of the multi-objective evolutionary algorithm and the multi-objective algorithm based on the AP clustering algorithm,the improved multi-objective evolutionary algorithm in this paper performs better overall.Therefore,the algorithm in this paper can be used for more accurate detection of complex network communities.In order to accurately cluster real data sets,an improved multi-objective evolution clustering algorithm is proposed.The algorithm uses multi-objective evolution as the framework.First,the spectral clustering algorithm is used to transform the sample features,and then the fuzzy C-means(FCM)clustering algorithm and the k-means(kmeans)clustering algorithm are fused to the initial clustering center.As the independent variable of the iteration;adding the reordering operation and the fusion result of the ranking results of the FCM and k-means clustering results in the calculation of individual fitness to ensure the diversity of the clustering results and the uniformity among the categories;in addition,Using the best individuals of the contemporary and the best individuals of the history to cross the other individuals of the contemporary,form new generations of the next generation,and ensure the evolutionary trend of the group;Finally,use the hill-climbing algorithm to improve the multi-objective evolution algorithm to quickly find Optimal solution.Experiments on UCI dataset and artificial dataset show that the algorithm has high accuracy.