Link prediction method based on complex network dynamics model

Abstract

Link prediction is an important part of the study of missing links and future formations in complex networks.Currently,network structure-based link prediction methods are rich in results.Research on link prediction based on complex network dynamics model is rare.Firstly,a complex network dynamics model for unlicensed and undirected networks was constructed.Then the quantitative evaluation index of the link prediction node centrality based on the complex network dynamics model was given.Finally,the link prediction method defined by the complex network dynamics model was proposed by the given node centrality quantitative index.Experiments on real network datasets show that the proposed link prediction method has obvious prediction accuracy improvement.

Keywords

• complex network;link prediction;network dynamics