Link prediction based on combined influence and effective path

Yang Tian,Han Li,Hui Tian,Xuzhen Zhu

doi:10.1142/s0217979219502497

Abstract

Link prediction based on topological similarity in complex networks obtains more and more attention both in academia and industry. Most researchers believe that two unconnected endpoints can possibly make a link when they have large influence, respectively. Through profound investigations, we find that at least one endpoint possessing large influence can easily attract other endpoints. The combined influence of two unconnected endpoints affects their mutual attractions. We consider that the greater the combined influence of endpoints is, the more the possibility of them producing a link. Therefore, we explore the contribution of combined influence for similarity-based link prediction. Furthermore, we find that the transmission capability of path determines the communication possibility between endpoints. Meanwhile, compared to the local and global path, the quasi-local path balances high accuracy and low complexity more effectually in link prediction. Therefore, we focus on the transmission capabilities of quasi-local paths between two unconnected endpoints, which is called effective paths. In this paper, we propose a link prediction index based on combined influence and effective path (CIEP). A large number of experiments on 12 real benchmark datasets show that in most cases CIEP is capable of improving the prediction performance.

Highlights

In the past few years, many researchers explored the potential or missing links in complex networks,[1,2] which can be described with nodes and relations between two unconnected endpoints
Researchers contribute to link prediction and obtain fruitful achievements, especially the findings based on the topological similarity of the complex networks
With respect to topological similarity, the traditional researches ignore the combined influence of endpoints and the transmission capabilities of quasi-local paths which can enhance the link prediction accuracy

Summary

Introduction

In the past few years, many researchers explored the potential or missing links in complex networks,[1,2] which can be described with nodes and relations between two unconnected endpoints. To optimize the low prediction accuracy of the local similarity indices and the high computational complexity of the global similarity indices, researchers propose many quasi-local similarity indices.[30,31] For example, Local Path index (LP)[25,30] considers the number of all two-step and three-step paths with the number of all two-step paths preferred. Yao et al.[33] propose a Resources from Short Paths index (RSP) It considers the interactions among the paths with different length and adopts the resource-traffic flow mechanism to measure the interactions among the paths between node pairs. It uses information entropy to measure the contribution of a path in link prediction and further quantifies the contribution of a path with both

Methods

Results

Conclusion