Research map / Graph learning / Graph self-supervised learning
Mitigating Noise and Imbalance in Social Governance Graphs for Multi-Type Risk Assessment
Di Jin, Haotian Zhao, Xiaobao Wang, Fengyu Yan, Dongxiao He
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 |
|---|---|---|---|
| Use HeCoGNN instead of RGCN | HeCoGNN · SG-I Person nodes | Macro-F1: 0.6874 -> 0.7471 | Improved |
| Use HeCoGNN instead of RGCN | HeCoGNN · SG-I Location nodes | Macro-F1: 0.6898 -> 0.8058 | Improved |
| Use HeCoGNN instead of RGCN | HeCoGNN · SG-I Person nodes | AUC: 0.8600 -> 0.8452 | Hurt |
| Use HeCoGNN instead of MLP | HeCoGNN · SG-II Event nodes | Macro-F1: 0.6791 -> 0.6774 | No clear effect |
| Use HeCoGNN instead of RGCN | HeCoGNN · SG-II Person nodes | Macro-F1: 0.4962 -> 0.6310 | Improved |
| Use HeCoGNN instead of RGCN | HeCoGNN · ACM | Macro-F1: 0.8325 -> 0.8324 | No clear effect |
| Use HeCoGNN instead of RGCN | HeCoGNN · ACM | Micro-F1: 0.8305 -> 0.8416 | Improved |
| Add GateNet edge gating to ContextNet-only model (ablation) | HeCoGNN · SG-I | Accuracy | Improved |
| Add ContextNet attention to GateNet-only model (ablation) | HeCoGNN · SG-I | Accuracy | Improved |
| Use MLP gating plus dynamic attention instead of simple or fixed-weight variants (ablation) | HeCoGNN · Social governance graphs | Macro-F1/AUC | Improved |
| Add graph structure via HGT/RGCN over MLP for event nodes | MLP · Social risk case graph (event nodes) | Accuracy | No clear effect |
Closest papers by meaning
- IGL-Bench: Establishing the Comprehensive Benchmark for Imbalanced Graph Learning (ICLR 2025 · similarity 0.82)
- Dissecting the Failure of Invariant Learning on Graphs (NeurIPS 2024 · similarity 0.82)
- TEG-DB: A Comprehensive Dataset and Benchmark of Textual-Edge Graphs (NeurIPS 2024 · similarity 0.82)
- IOHunter: Graph Foundation Model to Uncover Online Information Operations (AAAI 2025 · similarity 0.82)
- Context-aware Graph Neural Network for Graph-based Fraud Detection with Extremely Limited Labels (AAAI 2025 · similarity 0.81)
- TGB 2.0: A Benchmark for Learning on Temporal Knowledge Graphs and Heterogeneous Graphs (NeurIPS 2024 · similarity 0.81)
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.
