Research map / Graph learning / Knowledge and temporal graphs
Distilling Meta Knowledge on Heterogeneous Graph for Illicit Drug Trafficker Detection on Social Media
Yiyue Qian, Yiming Zhang, Yanfang Ye, Chuxu 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 MAML meta-learning (vs 4-layer DNN) (ablation) | MetaHG · Instagram drug trafficker, 1-shot | ACC: 0.7247 -> 0.8489 | Improved |
| add MAML meta-learning (vs 4-layer DNN) (ablation) | MetaHG · Instagram drug trafficker, 20-shot | ACC: 0.7932 -> 0.9354 | Improved |
| add knowledge distillation (ablation) | MetaHG · Instagram drug trafficker, 1-shot | ACC: 0.7914 -> 0.8489 | Improved |
| add knowledge distillation (ablation) | MetaHG · Instagram drug trafficker, 20-shot | ACC: 0.8804 -> 0.9354 | Improved |
| add graph structure refinement (ablation) | MetaHG · Instagram drug trafficker, 1-shot | ACC: 0.8078 -> 0.8489 | Improved |
| add graph structure refinement (ablation) | MetaHG · Instagram drug trafficker, 20-shot | ACC: 0.8934 -> 0.9354 | Improved |
| add self-supervised module in R-GCNs (ablation) | MetaHG · Instagram drug trafficker, 1-shot | ACC: 0.8147 -> 0.8489 | Improved |
| add self-supervised module in R-GCNs (ablation) | MetaHG · Instagram drug trafficker, 20-shot | ACC: 0.9056 -> 0.9354 | Improved |
| increase SSL loss weight beyond 5 (ablation) | MetaHG · Instagram drug trafficker | performance | Hurt |
| MetaHG vs R-GCNs+MAML (best baseline) | meta-learning on heterogeneous graph · Instagram drug trafficker, 1-shot | ACC: 0.7853 -> 0.8489 | Improved |
Closest papers by meaning
- T-NET: Weakly Supervised Graph Learning for Combatting Human Trafficking (AAAI 2024 · similarity 0.85)
- Unveiling the Threat of Fraud Gangs to Graph Neural Networks: Multi-Target Graph Injection Attacks Against GNN-Based Fraud Detectors (AAAI 2025 · similarity 0.81)
- DiG-In-GNN: Discriminative Feature Guided GNN-Based Fraud Detector against Inconsistencies in Multi-Relation Fraud Graph (AAAI 2024 · similarity 0.81)
- DiffGAD: A Diffusion-based Unsupervised Graph Anomaly Detector (ICLR 2025 · similarity 0.80)
- Context-aware Graph Neural Network for Graph-based Fraud Detection with Extremely Limited Labels (AAAI 2025 · similarity 0.80)
- Dynamic Neighborhood Modeling via Node-Subgraph Contrastive Learning for Graph-Based Fraud Detection (AAAI 2025 · similarity 0.80)
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.
