Research on the Influence of Network Motif on Link Prediction

Abstract

Link prediction is an important means to reveal the evolution of the network, which originated in the field of computer. Currently the link prediction algorithm is mostly based on the local characteristics of the network. Moreover, the motif is just a new concept and method for the research of the local structure of the complex network. In this paper, we studied the effect between the motifs and link prediction by the Rand- ESU algorithm and common neighbors Index and found that it has an important influence. We also analyzed and validated the mechanism of the interaction, so as to provide a strong basis to design the efficiently link prediction algorithm and enhance the understanding of the network.