The Research On Algorithms Of Social Network Community Discovery Based On Similarity Of Nodes

Currently,social networks feature a high theoretical and applicable value.The community structure has the characteristics of dense connections between internal nodes of the community and spare connections between the communities.In different fields of science,researchers have proposed different interpretations of the definition of community structure and the meaning of community structure,which results in a series of algorithms for community discovery with different angles and strategies.However,how to reduce the time complexity of the algorithm and improve the accuracy and stability of the algorithm remains to be solved.This article will conduct the related research on the following 3 aspects.(1)To explain the concept and characteristics of social networks,and then elaborate on the different network models of social networks.The traditional community discovery algorithms will be introduced and classified.The advantages,disadvantages,and performance of each algorithm will be introduced specifically.Based on the research of community discovery on algorithms and the latest node similarity,two revolutionary methods based on similarity of nodes will be proposed.(2)By referring to the number of clusters in the community,the number of common neighbors between nodes,while referring to the shortest path length between nodes,a new community discovery algorithm based on node similarity applies.Then we should compare the current community structure discovery algorithm.To be more specific,the algorithm does not limit the conditions,which can be also converged at the same time,the demand for the amount of network information is not so necessary.However,the community has a higher degree of accuracy.Comparing the network simulation results with several different community algorithms,the algorithm can be accurate and efficient.(3)A multi-level node similarity calculation method is proposed,which can effectively calculate the similarity between nodes.On the other hand,it can solve the problem of node merger selection when the similarity of nodes is mostly equal.Based on this node,similarity calculation method,the associative tightness metric between groups,and the community discovery model will be built.Also,the community classification experiment should be conducted on the real social network.Inspired by the results of the experimental analysis with other classical community discovery algorithms,this model can be successfully applied to an accurate determination of community members.