Mantra Name
Description
Mantra Name
Dataset Overview
Lifting Methods
Structural-based Liftings
- Cellular: Cycle-based lifting
- Simplicial: Clique complex lifting
- Hypergraph: k-hop lifting
Feature Lifting
- Projected sum
Model Performance
| Model | Accuracy (%) | Std Dev (±) |
|---|---|---|
| GCN | 42.14 | 2.72 |
| GAT | 18.09 | 0.65 |
| GIN | 76.14 | 0.14 |
| SCN | 79.48 | 1.36 |
| SCCNN | 95.08 | 0.56 |
| SaNN | 81.76 | 1.37 |
| GCCN | 86.76 | 1.27 |
| HOPSE-M | 91.50 | 1.45 |
| HOPSE-G | 81.75 | 1.26 |
Key Insights
- SCCNN achieves the best performance with 95.08% accuracy
- HOPSE-M and GCCN also perform strongly, with accuracies above 86%
- GAT shows the lowest performance (18.09%)
- GCN and GAT have the highest variability (std dev 2.72 and 0.65, respectively)
Reproducibility
to be released