AI Paper Atlas

Research map / Graph learning / Graph self-supervised learning

GADBench: Revisiting and Benchmarking Supervised Graph Anomaly Detection

Jianheng Tang, Fengrui Hua, Ziqi Gao, Peilin Zhao, Jia Li

NeurIPS 2023 · Graph learning › Graph self-supervised learning · approach: Graph anomaly detection · 19 citations (OpenAlex)

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 changedWhereMeasuredEffect
Use tree ensembles with simple neighborhood aggregation instead of GNN methodsSupervised graph anomaly detection · 10 real-world GAD datasets, fully-supervisedAUPRC: +12.9% AUPRC vs BWGNNImproved
Use tree ensembles with simple neighborhood aggregation instead of GNN methodsSupervised graph anomaly detection · 10 real-world GAD datasets, semi-supervisedAUPRC: +8.0% AUPRC vs GHRNImproved
Tune model hyperparameters with random searchStandard GNNs · GADBench datasetsAUPRC: GraphSAGE average AUPRC improvement of 10.4%Improved
Tune specialized GNN hyperparametersSpecialized GNNs · GADBench datasetsAUPRCImproved
Add Neighborhood Averaging (ablation)XGBoost · 10 datasets, semi-supervisedAUPRC: 37.5% -> 38.9%Improved
Add Neighborhood Averaging (ablation)XGB-Graph · 10 datasets, semi-supervisedAUPRC: 42.91 -> 44.30Improved
Add Neighborhood Averaging (ablation)XGB-Graph · 10 datasets, fully-supervisedAUPRC: 61.09 -> 59.58Hurt
Add Neighborhood Averaging (ablation)RF-Graph · 10 datasets, semi-supervisedAUPRC: 43.37 -> 45.14Improved
Add Neighborhood Averaging (ablation)RF-Graph · 10 datasets, fully-supervisedAUPRC: 59.65 -> 58.54Hurt
Add Neighborhood Averaging (ablation)GIN · 10 datasets, semi-supervisedAUPRC: 30.26 -> 33.91Improved
Increase neighbor aggregation depth from 0 to 2 layers (ablation)XGB-Graph and RF-Graph · Most datasetsperformanceImproved
Increase neighbor aggregation depth beyond 2 layers (ablation)XGB-Graph and RF-Graph · Most datasetsperformanceNo clear effect

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy