Study On Extracting Backbones Of Complex Networks Based On Information Diffusion

Complex network is a useful tool to study and analyze the real world networks.However,the growing number of links and nodes brings a challenge for us to analyze the complex networks.How to improve the efficiency of extracting backbone has gradually been in a hot research.With the coming of information society,information diffusion theories have been widely applied in the studies on network science.At the same time,the rapid development of the Internet provides data support for the studies on diffusion of mass information among networks and a new point of extracting backbones of complex networks.In this paper,we firstly review the existing methods of identifying backbones of complex networks;and we focus on the diffusion rales and the parameters of the common information diffusion models.On the basis of this,we put forward a backbone extraction model of complex network which is based on information diffusion.At the first,the model identifies the critical nodes of a network by the methods of evaluation of node importance;and secondly,during the simulation of SI model,these nodes will be the initial ones to spread the information.At the last,according to the performance of the links during the information diffusion,the model could recognize the critical links to construct the backbone of the complex network.In order to ensure the effectiveness and steadiness,this model chooses five different methods to identify the significant nodes,and separately design evaluation indicators from the perspective of node and link.The results of this study show that this model could be useful to understand the structure of network from the point of information diffusion,and identify the critical links during information spread.What's more,it illustrates the information could be transmitted effectively in the way of backbone.