Graph Neural Network for Product Recommendation on the Amazon Co-purchase Graph
This work addresses the challenge of identifying relevant information in large-scale data for product recommendation systems, though it is incremental as it compares existing methods on a known dataset.
The study evaluated four Graph Neural Network architectures (LightGCN, GraphSAGE, GAT, PinSAGE) on the Amazon Product Co-purchase Network for link prediction, analyzing trade-offs in performance, scalability, training complexity, and generalization to inform real-world recommendation deployment.
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.