Predicting mortality in critically ill patients with diabetes using machine learning and clinical notes

Table 2 Performance of machine learning models

	AUC	PPV	TPR	F1 score
Logistic regression	0.82	0.63	0.25	0.35
Random forest	0.86	0.81	0.34	0.48
AdaBoost	0.84	0.68	0.32	0.44
Gradient boosting	0.83	0.55	0.39	0.46
XGBoost	0.87	0.77	0.37	0.50
ANN	0.86	0.77	0.34	0.47
Majority voting	0.87	0.82	0.33	0.47

AUC area under the curve, PPV positive predictive value, TPR true positive rate

ISSN: 1472-6947