Similarity Index Of Hypernetwork And Its Application

With the rapid development of emerging industries such as artificial intelligence and big data,the research of network science is more and more in-depth and very important.For example,the similarity index of complex networks is widely used in the fields of link prediction,recommendation technology and data mining.In recent years,hypernetwork based on Hypergraph structure has been paid more and more attention because it is more in line with the characteristics of real network structure,and a lot of research has been done on hypernetwork structure.This paper systematically studies and analyzes the similarity indicators on the hypernetwork,puts forward four types of similarity indicators from the two aspects of hyperedge similarity and node similarity,evaluates and analyzes the prediction accuracy of similarity indicators on the Small-world hypernetwork model,and gives the concept of hypernetwork difference index in Xining public transportation hypernetwork,Combined with the capacity load model of non-uniform load distribution,the importance of hyperedge and node is analyzed and studied.The specific research contents of this paper include:(1)Based on Hypergraph structure,four kinds of similarity indexes of hypernetworks are proposed.Based on the similarity index based on common neighbor and path,combined with the hyperedge and node characteristics of hypergraph structure,this paper improves and proposes four kinds of hypernetwork similarity index based on common hyperneighbor,preferred connection,local hyperpath and HKatz;(2)Build Small-world hypernetwork models and study the similarity index.In this paper,four kinds of Small-world hypernetwork models such as uniform reconnection / edging and non-uniform reconnection / edging are constructed,and the simulation experiments are carried out by using the model evaluation index AUC in the three kinds of typical hypernetwork models.The results show that the accuracy of the improved HKatz index is obviously better than other indexes;(3)Construct Xining bus hypernetwork model and study and apply the similarity index.Based on the real data set,this paper constructs the Xining bus hypernetwork model,on which AUC is used to evaluate the accuracy of the hypernetwork similarity index.In addition,the HKatz index is used to propose the difference index of the hypernetwork,and the simulation experiment of Xining bus hypernetwork is carried out in combination with the cascading failure models of the hypernetwork.The results show that,From the perspective of transportation means that passengers can choose,as well as the cost of bus lines and bus stops in case of natural disasters or damage,maintenance and shutdown,the bus lines and bus stops with greater differences are more important.