Table 9. Results of model validation on node clustering task for random graphs (both metrics lie between (−1,1) and higher value means better results.
Bold corresponds to the best metric for each dataset.
| HOPE | WALKLETS | |||
|---|---|---|---|---|
| Modularity | Silhouette | Modularity | Silhouette | |
| CORA (Original) | 0.1222 | 0.2593 | 0.7353 | 0.0812 |
| CORA (Barabási-Albert) | 0.0005 | 0.1807 | 0.1465 | 0.0046 |
| CORA (Erdős-Rényi) | −0.0022 | 0.0216 | 0.0184 | 0.0080 |
| CORA (Watts-Strogatz) | 0.0629 | 0.1180 | 0.5251 | 0.0212 |
| CITESEER (Original) | 0.1748 | 0.5492 | 0.7263 | 0.0566 |
| CITESEER (Barabási-Albert) | −0.0008 | 0.3731 | 0.0397 | −0.0006 |
| CITESEER (Erdős-Rényi) | 0.0031 | 0.1495 | −0.0040 | 0.0085 |
| CITESEER (Watts-Strogatz) | 0.0344 | 0.1270 | 0.4941 | 0.0155 |