Research And Application Of Community Discovery Algorithm Based On Triangular Phantom

With the development of complex networks,especially in the rapid development of the Internet under the social network,a number of new and important discoveries have emerged.,attracting a large number of researchers in various disciplines into it.Community structure as one of its ubiquitous topological characteristics,for its exploration is helpful to reveal the relationship between network structure and function,so that the community structure in the network has important theoretical and practical significance.In this paper,we focus on the research of existing community discovery algorithms.In this paper,introduces and analyzes the existing edge based representation of the network community discovery algorithm for computing the bottleneck problem,we design a community structure discovery based on triangular motifs and expectation-maximization model and algorithm.The main work of this paper is as follows:(1)In order to solve the bottleneck problem of the existing community discovery algorithm using edge representation networks,this paper design a community discovery algorithm based on triangular motif.In this paper,the observation network is represented by a triangular motif,which simplifies the network structure hypothesis and provides a basis for designing an effective algorithm.Based on the modeling of the formation process of the triangular motif,a model of community structure discovery based on triangle motif and expectation maximization is designed,in order to simulate the observation network,and put forward the corresponding community finding algorithm,the triangular phantom and on both sides of triangle motif adopted in the algorithm as the calculation object,while ensuring the real network structure,by reducing the calculation object to improve the efficiency of the algorithm,the link probability return node association membership and mixed between communities,and the experiment results show that the proposed algorithm is feasible and effective in this paper.(2)In view of the efficiency of community discovery algorithm based on CSDTME model,this paper designs an improved algorithm based on CSDTME model.Based on CSDTME model of community discovery algorithm in time efficiency has been greatly improved on,but through the study of the algorithm further,can be found in time and space can also be further improved the algorithm can find the community structure of the network more quickly.The CSDTME model of community discovery algorithm in time and space in the process of solving the parameters based on the problems,coping strategies were put forward in space and time,mainly by reducing the number of iterations and the intermediate variable storage,to provide the efficiency of the algorithm.Finally through the experiment proved that the improved CSDTME model based on community discovery algorithm is feasible and effective.And based on this algorithm,the design and implementation of a community classification system based on CSDTME model is simple.