Table 2.

Single-model and ensemble performance on the test set.

Model	Accuracy	Precision	Recall	Specificity	F1-Score	ROC-AUC	PR-AUC	Brier score
Logistic Regression	0.8777	0.4833	0.9062	0.874	0.6304	0.9486	0.7417	0.1014
Random Forest	0.9209	0.6786	0.5938	0.9634	0.6333	0.9512	0.6422	0.0592
XGBoost	0.9137	0.6053	0.7188	0.939	0.6571	0.951	0.6327	0.069
LightGBM	0.9209	0.625	0.7812	0.939	0.6944	0.9605	0.6769	0.0596
Gradient Boosting	0.9137	0.625	0.625	0.9512	0.625	0.947	0.653	0.0613
SVM	0.9029	0.549	0.875	0.9065	0.6747	0.9416	0.5441	0.0662
Neural Network	0.8993	0.8333	0.1562	0.9959	0.2632	0.8706	0.6457	0.0698
AdaBoost	0.9209	0.6316	0.75	0.9431	0.6857	0.8982	0.7367	0.2252
Ensemble (Voting)	0.9317	0.6757	0.7812	0.9512	0.7246	0.9599	0.6927	0.0571