Research On Community Structure Detection Algorithm Based On Node Similarity

In view of the problems of large amount of calculation,low accuracy and instability of algorithm result,we have put forward a new community structure detection algorithm of local information based on the Newman fast algorithm and LPAm algorithm by depicting their strengths combined.The algorithm uses node degrees and shared neighbor numbers to define nodes similarity,and divides the network into initial community structure by node similarity and a similarity threshold default at first and then according to another preset number of communities parameter threshold to further optimize the community structure,Finally,getting a higher accuracy results of divided communities.Algorithm performance analysis has showed that the algorithm has linear time complexity.The results have showed that the proposed algorithm performs well than the GN algorithm and Newman fast algorithm and LPAm algorithm in terms of accuracy,and it'is a stability algorithm.And it has good results for large networks of community structure prediction,therefore,having better usability.