Just Propagate: Unifying Matrix Factorization, Network Embedding, and LightGCN for Link Prediction

TL;DR

This paper introduces a unified framework for link prediction that encompasses matrix factorization, network embedding, and LightGCN, providing insights into their design factors and guiding future research.

Contribution

It unifies various link prediction models into a single framework and offers analysis that deepens understanding and inspires new methods.

Findings

Key design factors identified within the unified framework

Analysis reveals how different models relate and differ

Results suggest directions for future link prediction research

Abstract

Link prediction is a fundamental task in graph analysis. Despite the success of various graph-based machine learning models for link prediction, there lacks a general understanding of different models. In this paper, we propose a unified framework for link prediction that covers matrix factorization and representative network embedding and graph neural network methods. Our preliminary methodological and empirical analyses further reveal several key design factors based on our unified framework. We believe our results could deepen our understanding and inspire novel designs for link prediction methods.