A Survey on Hyperlink Prediction

TL;DR

This survey comprehensively reviews hyperlink prediction in hypergraphs, categorizing methods and benchmarking their performance across applications, highlighting the dominance of deep learning approaches.

Contribution

It introduces a new taxonomy for hyperlink prediction methods and provides a comparative benchmark study across different categories and applications.

Findings

Deep learning methods outperform others in hyperlink prediction.

The survey offers a systematic classification of existing methods.

Benchmark results demonstrate the effectiveness of deep learning approaches.

Abstract

As a natural extension of link prediction on graphs, hyperlink prediction aims for the inference of missing hyperlinks in hypergraphs, where a hyperlink can connect more than two nodes. Hyperlink prediction has applications in a wide range of systems, from chemical reaction networks, social communication networks, to protein-protein interaction networks. In this paper, we provide a systematic and comprehensive survey on hyperlink prediction. We propose a new taxonomy to classify existing hyperlink prediction methods into four categories: similarity-based, probability-based, matrix optimization-based, and deep learning-based methods. To compare the performance of methods from different categories, we perform a benchmark study on various hypergraph applications using representative methods from each category. Notably, deep learning-based methods prevail over other methods in hyperlink prediction.