Research map / Graph learning / Knowledge and temporal graphs
Rel-MOSS: Towards Imbalanced Relational Deep Learning on Relational Databases
Jun Yin, Peng Huo, Bangguo Zhu, Hao Yan, Senzhang Wang, Shirui Pan, Chengqi Zhang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| add Rel-Syn minority synthesizer (ablation) | Rel-MOSS · f1-driver-top3 | balanced accuracy: 0.5627 -> 0.8098 | Improved |
| add Rel-Syn minority synthesizer (ablation) | Rel-MOSS · avito-user-clicks | balanced accuracy: 0.5000 -> 0.6221 | Improved |
| add Rel-Syn minority synthesizer (ablation) | Rel-MOSS · stack-user-engagement | balanced accuracy: 0.5745 -> 0.8112 | Improved |
| add Rel-Syn minority synthesizer (ablation) | Rel-MOSS · hm-user-churn | balanced accuracy: 0.5956 -> 0.5736 | Hurt |
| add Rel-Gate gating controller (ablation) | Rel-MOSS · f1-driver-top3 | balanced accuracy: 0.7651 -> 0.8098 | Improved |
| add Rel-Gate gating controller (ablation) | Rel-MOSS · stack-user-engagement | balanced accuracy: 0.7903 -> 0.8112 | Improved |
| add Rel-Gate gating controller (ablation) | Rel-MOSS · amazon-user-churn | balanced accuracy: 0.6043 -> 0.6363 | Improved |
| add Rel-Gate gating controller (ablation) | Rel-MOSS · f1-driver-dnf | balanced accuracy: 0.6623 -> 0.6510 | Hurt |
| add Rel-Gate gating controller (ablation) | Rel-MOSS · trial-study-outcome | balanced accuracy: 0.6273 -> 0.6252 | Hurt |
| add Rel-MOSS over native RDL on severely imbalanced data | RDL · f1-driver-top3 | balanced accuracy: 0.5000 -> 0.8098 | Improved |
| add Rel-MOSS over native RDL | RDL · avito-user-visits | balanced accuracy: 0.5000 -> 0.6298 | Improved |
| add classic imbalance techniques (SMOTE) to RDL on balanced-ish dataset | RDL · amazon-user-churn | balanced accuracy: 0.6309 -> 0.6158 | Hurt |
Closest papers by meaning
- Position: Relational Deep Learning - Graph Representation Learning on Relational Databases (ICML 2024 · similarity 0.89)
- RelGNN: Composite Message Passing for Relational Deep Learning (ICML 2025 · similarity 0.88)
- Griffin: Towards a Graph-Centric Relational Database Foundation Model (ICML 2025 · similarity 0.85)
- Joint Relational Database Generation via Graph-Conditional Diffusion Models (NeurIPS 2025 · similarity 0.85)
- Relational Learning in Pre-Trained Models: A Theory from Hypergraph Recovery Perspective (ICML 2024 · similarity 0.84)
- WikiDBs: A Large-Scale Corpus Of Relational Databases From Wikidata (NeurIPS 2024 · similarity 0.84)
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.
