Research Of Community Division Of Bipartite Network Based On PageRank Algorithm

With the development of the network, posting online is becoming increasingly popular in our daily life. The community division of netizens and posts, which is according to posting online, has a great significance to the judgment of Internet mercenaries and spams. Netizens and posts constitute a bipartite network, which is one of the manifestations of complex networks. Bipartite division of the network is a branch of complex networks.The method of community detection in bipartite networks started with projection, it projected the number of the network's node type into one single and then makes full useof the method which is so mature in single networks to detect the community.The first chapter summaries the related knowledge of complex networks and the current research, states the statistical description of complex networks in everyday applications. The second chapter introduces the complex networks and their property. The third chapter introduces the PageRank algorithm, the thought of random walk and Markov chains. The forth chapter introduces the algorithm of bipartite network community division based on the PageRank algorithm, and reveals its experimental conclusion in the classical networks. The final section of this chapter describes the evaluation criteria of community division- modularity Q function. The final chapter is conclusion. The algorithm in this study comes to a conclusion and further study of division community is given.This paper presents a community unilateral nodes of bipartite network divided clustering algorithm based on the PageRank algorithm and modularity of information in the network. It simulates the information spreading. The amount of information received is as accordance for the merging among the communities. Leading in the module as the basis to judge the community division. It firstly transfers the bipartite network into transition probability matrix, then using the relationship of the nodes to conduct probability transition, finally gets the relationship between the probabilities of unilateral node metastasis, and take it to bipartite network community division. Finally, using actual network test the performance of the algorithm.Result shows that the algorithm can be accurately used into bipartite network community division, and can also get community division of higher quality.