Table 4 Results of the feature ablation study in CrossFeat
Method | Binary classification | Regression | ||
|---|---|---|---|---|
AUROC | AUPRC | RMSE | MAE | |
\(\hbox {Drug}_{\textrm{Fingerprint}}\) | 0.802 ± 0.00 | 0.806 ± 0.01 | 0.88 ± 0.08 | 0.65 ± 0.04 |
\(\hbox {Drug}_{\textrm{Mol2vec}}\) | 0.802 ± 0.00 | 0.812 ± 0.01 | 0.86 ± 0.02 | 0.64 ± 0.02 |
Side-\(\hbox {Effect}_{\textrm{Semantic}}\) | 0.797 ± 0.01 | 0.805 ± 0.02 | 0.88 ± 0.06 | 0.67 ± 0.02 |
Side-\(\hbox {Effect}_{\textrm{Word}}\) | 0.804 ± 0.01 | 0.811 ± 0.01 | 0.85 ± 0.05 | 0.65 ± 0.03 |