Table 2.

Comparative performance metrics across machine learning models and validation sets.

Model	Dataset	Accuracy mean	Accuracy std	F1 mean	F1 std	ROC AUC mean	ROC AUC std	Precision	Recall
CatBoost	Cross-validation	0.918	0.015	0.892	0.014	0.923	0.018	0.875	0.912
CatBoost	Test	0.905	0.014	0.881	0.013	0.911	0.017	0.862	0.901
CatBoost	External Val.	0.875	0.016	0.852	0.015	0.892	0.017	0.833	0.872
Random Forest	Cross-validation	0.832	0.016	0.812	0.014	0.853	0.018	0.802	0.823
Random Forest	Test	0.845	0.015	0.828	0.013	0.864	0.017	0.815	0.842
Random Forest	External Val.	0.812	0.016	0.792	0.014	0.838	0.017	0.783	0.801
Gradient Boost	Cross-validation	0.824	0.019	0.794	0.017	0.842	0.021	0.782	0.807
Gradient Boost	Test	0.837	0.018	0.812	0.016	0.853	0.020	0.798	0.827
Gradient Boost	External Val.	0.803	0.019	0.774	0.017	0.828	0.020	0.762	0.787
K Neighbors	Cross-validation	0.854	0.013	0.841	0.015	0.881	0.015	0.832	0.851
K Neighbors	Test	0.868	0.012	0.854	0.014	0.892	0.014	0.843	0.865
K Neighbors	External Val.	0.842	0.013	0.824	0.015	0.873	0.015	0.813	0.836
Linear SVC	Cross-validation	0.826	0.017	0.803	0.013	0.848	0.019	0.795	0.812
Linear SVC	Test	0.835	0.016	0.815	0.012	0.857	0.018	0.805	0.825
Logistic Regression	Cross-validation	0.89	0.02	0.86	0.025	0.91	0.018	0.87	0.84
Logistic Regression	Test	0.85	0.03	0.86	0.014	0.86	0.035	0.95	0.75
Logistic Regression	External Val.	0.875	0.02	0.667	0.024	0.95	0.016	0.6	0.75