A benchmark of four GNN architectures (LightGCN, GraphSAGE, GAT, PinSAGE) for link prediction on the Amazon co-purchase graph, reporting trade-offs between accuracy, training cost, and scalability.
Deepgnn documentation: Link prediction with pytorch backend
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.IR 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Graph Neural Network for Product Recommendation on the Amazon Co-purchase Graph
A benchmark of four GNN architectures (LightGCN, GraphSAGE, GAT, PinSAGE) for link prediction on the Amazon co-purchase graph, reporting trade-offs between accuracy, training cost, and scalability.