English

Graph Neural Network for Product Recommendation on the Amazon Co-purchase Graph

Information Retrieval 2025-08-21 v1 Machine Learning

Abstract

Identifying relevant information among massive volumes of data is a challenge for modern recommendation systems. Graph Neural Networks (GNNs) have demonstrated significant potential by utilizing structural and semantic relationships through graph-based learning. This study assessed the abilities of four GNN architectures, LightGCN, GraphSAGE, GAT, and PinSAGE, on the Amazon Product Co-purchase Network under link prediction settings. We examined practical trade-offs between architectures, model performance, scalability, training complexity and generalization. The outcomes demonstrated each model's performance characteristics for deploying GNN in real-world recommendation scenarios.

Keywords

Cite

@article{arxiv.2508.14059,
  title  = {Graph Neural Network for Product Recommendation on the Amazon Co-purchase Graph},
  author = {Mengyang Cao and Frank F. Yang and Yi Jin and Yijun Yan},
  journal= {arXiv preprint arXiv:2508.14059},
  year   = {2025}
}

Comments

15 pages, 5 figures, preprint

R2 v1 2026-07-01T04:57:14.959Z