Table 6.

The performances of the models in the second scenario.

Model	AUROC	Brier score	Precision	Recall	F1 score
LoR	0.837 (0.801–0.863)	0.147	0.549	0.738	0.629
SVM	0.836 (0.803–0.862)	0.147	0.552	0.739	0.632
RF	0.845 (0.807–0.875)	0.142	0.577	0.758	0.655
NN	0.837 (0.800–0.862)	0.149	0.550	0.737	0.629
Ho-Pham et al.	0.828 (0.784–0.862)	0.150	0.506	0.826	0.627
OSTA	0.656 (0.621–0.711)	0.206	0.744	0.365	0.489