Development of decision tree classification algorithms in predicting mortality of COVID-19 patients

Table 5 Comparing the performance of logistic regression and decision tree algorithms to predict the COVID-19 mortality in testing dataset with model fitted to imbalanced data

Models	Confusion matrix				Sensitivity (Recall)	Specificity	Precision	Accuracy	F-score	AUC
	Predicted	Actual	Mortality	Survived
	Mortality		TP	FP
	Survived		FN	TN
Logistic			26 42	13 935	0.38	0.98	0.66	0.94	0.48	0.93
CART			51 17	89 859	0.75	0.91	0.36	0.90	0.49	0.92
C5.0			28 8	40 940	0.41	0.99	0.78	0.95	0.41	0.78
CHAID			28 40	12 936	0.41	0.98	0.70	0.95	0.52	0.87

ISSN: 1865-1380