A directional evolution control model for network

Abstract

In order to control the evolution direction of network to achieve certain goals, we proposed Directional Evolution Control Model for Network Based on Percolation Coefficient (PC-DECMN). Firstly, after the analysis of the spreading behavior based on SI model, we found that the influence scope was only related to the topology structure of network and initial source of infection. So the definition of percolation coefficient to measure the influence scope was proposed based on the transfer closure of adjacency matrix. Then, PC-DECMN was established based on the definition of percolation coefficient. Simulation experiment using the guarantee data of certain financial institution shows that, PC-DECMN can effectively control the evolution of guarantee network so that it can resist the risk better, compared with the natural evolved guarantee network.