Graph-level Anomaly Detection on MUTAG (AUROC, AUPRC, FPR95)
30.8FPR95LGKDE
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| LGKDE2025.05 | 30.8 | 91.63 | 96.75 | |
| LGKDECategory=GNN-based Deep Learning Methods2025.05 | 30.8 | — | — | |
| SIGNETCategory=GNN-based Deep Learning Methods2025.05 | 32 | — | — | |
| MUSECategory=GNN-based Deep Learning Methods2025.05 | 34.4 | — | — | |
| UniFORMCategory=GNN-based Deep Learning Methods2025.05 | 35.2 | — | — | |
| CVTGADCategory=GNN-based Deep Learning Methods2025.05 | 38.8 | — | — | |
| OCGTLCategory=GNN-based Deep Learning Methods2025.05 | 39.2 | — | — | |
| GLADCCategory=GNN-based Deep Learning Methods2025.05 | 39.2 | — | — | |
| InfoGraph+KDEType=Two-stage (Representation Learning + KDE)2025.05 | 44 | 79.77 | 88.36 | |
| OCGINCategory=GNN-based Deep Learning Methods2025.05 | 45.6 | — | — | |
| GLocalKDCategory=GNN-based Deep Learning Methods2025.05 | 48 | — | — | |
| GAE+KDEType=Two-stage (Representation Learning + KDE)2025.05 | 49.6 | 81.94 | 91 | |
| LGKDE2025.05 | 50.91 | 72.97 | 80.35 | |
| SIGNET2025.05 | 53.89 | 70.61 | 78.23 | |
| MUSE2025.05 | 54.56 | 70.12 | 77.89 | |
| UniFORM2025.05 | 56.12 | 69.56 | 77.34 | |
| CVTGAD2025.05 | 58.34 | 68.89 | 76.56 | |
| GLocalKD2025.05 | 61.12 | 67.45 | 74.23 | |
| OCGIN2025.05 | 71.23 | 57.12 | 65.67 | |
| WL-SVM2025.05 | 78.8 | 62.18 | 50.94 | |
| WL-SVMCategory=Graph Kernel + Detector2025.05 | 78.8 | — | — | |
| PK-iF2025.05 | 86 | 47.98 | 45.89 | |
| WL-iF2025.05 | 86 | 65.71 | 55.15 | |
| PK-iFCategory=Graph Kernel + Detector2025.05 | 86 | — | — | |
| WL-iFCategory=Graph Kernel + Detector2025.05 | 86 | — | — | |
| PK-SVM2025.05 | 88.4 | 46.06 | 47.8 | |
| PK-SVMCategory=Graph Kernel + Detector2025.05 | 88.4 | — | — |