Table 4.

Comparison of CNN-DDI with DDIMDL

Algorithm	ACC	AUPR	AUC	F1	Precision	Recall
CNN-DDI	0.8871	0.9251	0.9980	0.7496	0.8556	0.7220
DDIMDL	0.8852	0.9208	0.9976	0.7585	0.8471	0.7182
DDIMDL*	0.8865	0.9230	0.9976	0.7559	0.8513	0.7204

The single asterisk represent DDIMDL with features selected by our method. It can be concluded that the drug category is effective as a new feature type