A Fast Parallel Approach for Neighborhood-based Link Prediction by Disregarding Large Hubs

Abstract

Link prediction can help rectify inaccuracies in various graph algorithms, stemming from unaccounted-for or overlooked links within networks. However, many existing works use a baseline approach, which incurs unnecessary computational costs due to its high time complexity. Further, many studies focus on smaller graphs, which can lead to misleading conclusions. Here, we study the prediction of links using neighborhood-based similarity measures on large graphs. In particular, we improve upon the baseline approach (IBase), and propose a heuristic approach that additionally disregards large hubs (DLH), based on the idea that high-degree nodes contribute little similarity among their neighbors. On a server equipped with dual 16-core Intel Xeon Gold 6226R processors, DLH is on average 1019x faster than IBase, especially on web graphs and social networks, while maintaining similar prediction accuracy. Notably, DLH achieves a link prediction rate of 38.1M edges/s and improves performance by 1.6x for every doubling of threads.

arXiv comment17 pages, 12 figures, 1 table

References in corpus (11)

• Link Prediction in Complex Networks: A Survey
• Clustering and preferential attachment in growing networks
• Predicting Missing Links via Local Information
• Solving the apparent diversity-accuracy dilemma of recommender systems
• Vertex similarity in networks
• Evaluating Link Prediction Methods
• Stacking Models for Nearly Optimal Link Prediction in Complex Networks
• Experimental analyses on 2-hop-based and 3-hop-based link prediction algorithms
• Link Prediction in Social Networks: the State-of-the-Art
• Bounded link prediction for very large networks
• An Approach for Link Prediction in Directed Complex Networks based on Asymmetric Similarity-Popularity