Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Link Prediction in Bipartite Networks

Şükrü Demir İnan Özer, Günce Keziban Orman, Vincent Labatut

TL;DR

This work tackles link prediction in bipartite graphs, framing the problem with two node sets and a set of missing links. It evaluates 19 methods across three real datasets, spanning traditional scores, topological features, bipartite embeddings, and repurposed GCN-based recommender models. Key findings show that DiffRec, a GCN-based personalized recommender model, achieves high AUPR in sparse networks, while the Structural Perturbation Method remains a strong heuristic baseline; embedding-based and topological-feature methods generally underperform compared with supervised approaches. The study demonstrates the value of supervised graph representation learning tailored to bipartite link prediction and points to future directions involving ensembles and broader datasets.

Abstract

Bipartite networks serve as highly suitable models to represent systems involving interactions between two distinct types of entities, such as online dating platforms, job search services, or ecommerce websites. These models can be leveraged to tackle a number of tasks, including link prediction among the most useful ones, especially to design recommendation systems. However, if this task has garnered much interest when conducted on unipartite (i.e. standard) networks, it is far from being the case for bipartite ones. In this study, we address this gap by performing an experimental comparison of 19 link prediction methods able to handle bipartite graphs. Some come directly from the literature, and some are adapted by us from techniques originally designed for unipartite networks. We also propose to repurpose recommendation systems based on graph convolutional networks (GCN) as a novel link prediction solution for bipartite networks. To conduct our experiments, we constitute a benchmark of 3 real-world bipartite network datasets with various topologies. Our results indicate that GCN-based personalized recommendation systems, which have received significant attention in recent years, can produce successful results for link prediction in bipartite networks. Furthermore, purely heuristic metrics that do not rely on any learning process, like the Structural Perturbation Method (SPM), can also achieve success.

Link Prediction in Bipartite Networks

TL;DR

This work tackles link prediction in bipartite graphs, framing the problem with two node sets and a set of missing links. It evaluates 19 methods across three real datasets, spanning traditional scores, topological features, bipartite embeddings, and repurposed GCN-based recommender models. Key findings show that DiffRec, a GCN-based personalized recommender model, achieves high AUPR in sparse networks, while the Structural Perturbation Method remains a strong heuristic baseline; embedding-based and topological-feature methods generally underperform compared with supervised approaches. The study demonstrates the value of supervised graph representation learning tailored to bipartite link prediction and points to future directions involving ensembles and broader datasets.

Abstract

Bipartite networks serve as highly suitable models to represent systems involving interactions between two distinct types of entities, such as online dating platforms, job search services, or ecommerce websites. These models can be leveraged to tackle a number of tasks, including link prediction among the most useful ones, especially to design recommendation systems. However, if this task has garnered much interest when conducted on unipartite (i.e. standard) networks, it is far from being the case for bipartite ones. In this study, we address this gap by performing an experimental comparison of 19 link prediction methods able to handle bipartite graphs. Some come directly from the literature, and some are adapted by us from techniques originally designed for unipartite networks. We also propose to repurpose recommendation systems based on graph convolutional networks (GCN) as a novel link prediction solution for bipartite networks. To conduct our experiments, we constitute a benchmark of 3 real-world bipartite network datasets with various topologies. Our results indicate that GCN-based personalized recommendation systems, which have received significant attention in recent years, can produce successful results for link prediction in bipartite networks. Furthermore, purely heuristic metrics that do not rely on any learning process, like the Structural Perturbation Method (SPM), can also achieve success.
Paper Structure (10 sections, 5 equations, 3 figures, 2 tables)

This paper contains 10 sections, 5 equations, 3 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Methods
  3. Traditional Link Scores
  4. Link Prediction via Topological Features
  5. Embedding-based Link Prediction
  6. Personalized Recommendation Conversion
  7. Experiments and Results
  8. Experimental Setup, Performance Evaluation and Hyper-parameter Tuning
  9. Results
  10. Conclusion

Figures (3)

  • Figure 1: Representation of the link prediction method that relies on network topological features (Section \ref{['sec:LinkPredTopoFeat']}).
  • Figure 2: Representation of the link prediction method that relies on graph embeddings (Section \ref{['sec:embed-based']}).
  • Figure 3: Representation of the link prediction method based on the repurposing of item recommendation techniques (Section \ref{['sec:perso-recom-conv']}).