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

2025/08/10 by Mengyang Cao, Cao, Mengyang, Yi Jin +4
Computer Science · Decision Sciences · #Blockchain Technology Applications and Security #FOS: Computer and information sciences #Impact of AI and Big Data on Business and Society #Information Retrieval (cs.IR) #Machine Learning (cs.LG) #Sentiment Analysis and Opinion Mining

paper · pdf · doi:10.48550/arxiv.2508.14059

openalex publication_date 2025/08/10 · openalex created_date 2025/10/16 · openalex updated_date 2026/07/28

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.

Citations

• Evaluating Graph Neural Networks for Link Prediction: Current Pitfalls and New Benchmarking
• Recommending Related Products Using Graph Neural Networks in Directed Graphs
• RecBole: Towards a Unified, Comprehensive and Efficient Framework for Recommendation Algorithms
• LightGCN: Simplifying and Powering Graph Convolution Network for Recommendation
• LightGCN
• Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks
• Fast Graph Representation Learning with PyTorch Geometric
• AliGraph: A Comprehensive Graph Neural Network Platform
• Pitfalls of Graph Neural Network Evaluation
• node2vec: Scalable Feature Learning for Networks
• DeepWalk
• Matrix Factorization Techniques for Recommender Systems

Related

• Graph Neural Networks for Recommendation: Reproducibility, Graph Topology, and Node Representation
• Benchmarking Deep Neural Networks for Modern Recommendation Systems
• A Novel Evaluation Perspective on GNNs-based Recommender Systems through the Topology of the User-Item Graph
• Building a Recommendation System Using Amazon Product Co-Purchasing Network
• GNN-GMVO: Graph Neural Networks for Optimizing Gross Merchandise Value in Similar Item Recommendation