Research map / Graph learning / GNN architectures
Link Prediction Based on Graph Neural Networks
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Learn link heuristics with a GNN over local enclosing subgraphs rather than use predefined heuristics | SEAL · USAir, NS, PB, Yeast, C.ele, Power, Router, E.coli | AUC | Helped |
| Use graph-level GNN learning on enclosing subgraphs instead of WLNM's fully-connected network | SEAL versus WLNM · NS, PB, Yeast, Power, Router, E.coli | AUC: 98.61$\pm$0.49 -> \textbf{98.85}$\pm$0.47; 93.49$\pm$0.47 -> \textbf{94.72}$\pm$0.46; 95.62$\pm$0.52 -> \textbf{97.91}$\ | Helped |
| Add double-radius structural node labels rather than no labeling or naive labelings (ablation) | SEAL · Link prediction experiments | AUC | Helped |
| Inject sampled negative links when generating node embeddings (ablation) | SEAL with node embeddings · Link prediction experiments | AUC | Helped |
| Add node2vec embeddings to the SEAL node information matrix (ablation) | SEAL · Link prediction datasets | AUC | Mixed |
| Increase enclosing-subgraph hop count beyond h=2, up to h>=3 (ablation) | SEAL · Link prediction datasets | AUC | No clear effect |
Closest papers by meaning
- GNNs Meet Sequence Models Along the Shortest-Path: an Expressive Method for Link Prediction (NeurIPS 2026 · similarity 0.90)
- A Topological Perspective on Demystifying GNN-Based Link Prediction Performance (ICLR 2024 · similarity 0.89)
- Revisiting Link Prediction: a data perspective (ICLR 2024 · similarity 0.89)
- Structural Information Enhanced Graph Representation for Link Prediction (AAAI 2024 · similarity 0.88)
- Bridging Theory and Practice in Link Representation with Graph Neural Networks (NeurIPS 2025 · similarity 0.88)
- Mixture of Link Predictors on Graphs (NeurIPS 2024 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.