Link prediction in complex networks: a local na\"{\i}ve Bayes model

TL;DR

This paper introduces a local naive Bayes model for link prediction in complex networks, improving accuracy by considering the varying influence of different common neighbors, validated through experiments on real networks.

Contribution

The paper proposes a novel local naive Bayes approach that assigns different weights to common neighbors, enhancing link prediction accuracy over traditional methods.

Findings

The proposed model outperforms common-neighbor-based methods in accuracy.

Experiments on eight real networks demonstrate the effectiveness of the approach.

A case study on the US air transportation network illustrates practical application.

Abstract

Common-neighbor-based method is simple yet effective to predict missing links, which assume that two nodes are more likely to be connected if they have more common neighbors. In such method, each common neighbor of two nodes contributes equally to the connection likelihood. In this Letter, we argue that different common neighbors may play different roles and thus lead to different contributions, and propose a local na\"{\i}ve Bayes model accordingly. Extensive experiments were carried out on eight real networks. Compared with the common-neighbor-based methods, the present method can provide more accurate predictions. Finally, we gave a detailed case study on the US air transportation network.