The Community Structure Detection Algorithm Based On Node Dependence

Complex network as the abstraction and modeling of real word complex systems,the development of complex network provide good reference on complex relationships of real life.Through study structure of complex network,we can get the common features.Community structure is an important property to complex networks.Community is a set of nodes connected more closely in network.Community discovery is to find the potential of small interest groups in the network.Through research on community structure in the node even side relationship to forecast network potential behavior between nodes,so,the community structure discovery is of great theoretical and practical significance.In this paper,the related research of complex network community discovery question has carried on,related association discovery algorithm was introduced,based on analysis of algorithms proposed in this paper,we propose the algorithm based on node-dependence,the main work of the paper is as follows:(1)First on the relevant association discovery algorithm was introduced,focusing on local association discovery algorithm analysis,based on local information label propagation algorithm is widely used because of dividing faster,but the algorithm also exist instability,poor quality problem.In this paper,according to the shortcomings of the algorithm,the label propagation algorithm based on the node dependence is proposed,and the initialization strategy of the node is changed.Firstly,the nodes in the network are clustered according to the dependence of nodes,we can get initial community,and then with the aid of label propagation algorithm to adjust the network community,which changes the instability of the original algorithm,accelerate the algorithm convergence,and the algorithm is verified on the real network.(2)In the second part,based on the cooperative data collected between the patent cooperation enterprises,the patent cooperation network is constructed and the related structure characteristics of the network are analyzed,the evolution model of the network also proposed.And based on thealgorithm of node dependence of this paper,the community structure of patent cooperation network has been found.The significance of the community structure in reality also analyzed,further illustrates the advantages of this algorithm.At the same time,it also shows that the algorithm is good for the local connection of the network.