Graph Classification on CUNEIFORM
91.7AccuracyEGIN-E-epsilon
Evaluation results
| Method | Links | |
|---|---|---|
| EGIN-E-epsilonepsilon parameter=retained, validation=10-fold cross-validation2025.12 | 91.7 | |
| EGIN-Eepsilon parameter=not retained, validation=10-fold cross-validation2025.12 | 91.3 | |
| ALL-INTraining Scope=Trained on all datasets, Node Properties=Standard2026.05 | 91.17 | |
| ALL-INTraining Scope=Trained on all datasets, Node Properties=0 props2026.05 | 89.88 | |
| EGIN-epsilonepsilon parameter=retained, validation=10-fold cross-validation2025.12 | 89.3 | |
| EGIN-C-epsilonepsilon parameter=retained, validation=10-fold cross-validation2025.12 | 89 | |
| EGIN-Cepsilon parameter=not retained, validation=10-fold cross-validation2025.12 | 88.6 | |
| EGINepsilon parameter=not retained, validation=10-fold cross-validation2025.12 | 87.3 | |
| ALL-IN-SPECIALIZEDTraining Scope=Trained per dataset, Node Properties=Standard2026.05 | 87.2 | |
| ALL-IN-SPECIALIZEDTraining Scope=Trained per dataset, Node Properties=0 props2026.05 | 85.19 | |
| EGCNvalidation=10-fold cross-validation2025.12 | 77.5 | |
| GATvalidation=10-fold cross-validation2025.12 | 76.9 | |
| FGW SPAttributes=Vector, Kernel=Shortest Path2018.05 | 76.67 | |
| DGCNNvalidation=10-fold cross-validation2025.12 | 38 | |
| HOPPERKAttributes=Vector2018.05 | 32.59 | |
| PSCNAttributes=Vector, K=102018.05 | 25.19 | |
| PSCNAttributes=Vector, K=52018.05 | 24.81 | |
| GINvalidation=10-fold cross-validation2025.12 | 21.2 | |
| PROPAKAttributes=Vector2018.05 | 12.59 |