Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
974616 | Physica A: Statistical Mechanics and its Applications | 2015 | 8 Pages |
•We propose a method to predict link directions.•Our method is stable and robust in many kinds of directed networks.•We discuss the potential role of ground node in link direction prediction problem.
Link prediction in directed network is attracting growing interest among many network scientists. Compared with predicting the existence of a link, determining its direction is more complicated. In this paper, we propose an efficient solution named Local Directed Path to predict link direction. By adding an extra ground node to the network, we solve the information loss problem in sparse network, which makes our method effective and robust. As a quasi-local method, our method can deal with large-scale networks in a reasonable time. Empirical analysis on real networks shows that our method can correctly predict link directions, which outperforms some local and global methods.