Community Discovery And Visualization Of Complex Network Based On Fuzzy Pattern Recognition

With the rapid development of computer technology in recent years,complexities of networks and the amount of data increases year by year.As a result,visualization became more and more important in the analysis and research field of the complex network.In many of the visualization method,visualization on the basis of community found has become a trend,and good results have been achieved in the field.In this paper,the main research is about community discovery and visualization of the complex network.The main work includes three parts:Firstly,thoughts about fuzzy mathematics is tried to be used in finding complex networks’ communities.And then,dynamic partitioning of network community will be implementation,in order to show the network features better.In this paper,centricity of degree and closeness are chose to measure the importance of nodes.In the process of implementation.First fuzzy pattern recognition is used to decide core nodes about degree of importance.Second fuzzy pattern recognition is used to predicate community which other nodes should belong to.The division of community can be finished by two fuzzy pattern recognitions,which use direct method as the way to recognize nodes.Principle of maximum membership degree also be used in the processes.Secondly,two measures are taken to realize communities’ discovery by using thoughts about fuzzy mathematics.What different from the single measure is that two measures must be considered together,in the second process of fuzzy pattern recognition.So,indirect method of fuzzy pattern recognition and close degree are used to recognize nodes.At last,visualization scheme based on classification results is given in the paper,and a visual platform is made by using Processing language.Then Ring-Nested Algorithm which based on circular layout and Traction-Polymerized Algorithm which based on Force--Directed Algorithm are realized.As a result,dynamic display of community classification results will be realized,which is convenience to show network features.In one words,this paper not only gives the way to divide the complex network community which based on fuzzy pattern recognition by using fuzzy mathematics,but also designs the visualization algorithm to show the results of community classification results.The algorithms in this paper has practical,and also provides a new perspective for the theoretical study of complex network visual field.