Prediction of neonatal deaths in NICUs: development and validation of machine learning models

Table 4 Performance measures of the selected machine learning models on original, SMOTE-oversampled dataset1, SMOTE-oversampled dataset4 and ADASYN-oversampled dataset

Model	Data	Accuracy	Precision	Specificity	Sensitivity	F-score	AUC
RF	Original data	0.91	0.97	0.62	0.94	0.95	0.90
	SMOTE-oversampled dataset1	0.92	0.98	0.94	0.92	0.95	0.97
	SMOTE-oversampled dataset4	0.94	0.96	0.92	0.97	0.94	0.98
	ADASYN-oversampled dataset	0.96	0.99	0.92	0.99	0.95	0.96
ANN	Original data	0.93	0.95	0.43	0.98	0.97	0.93
	SMOTE-oversampled dataset1	0.91	0.94	0.84	0.94	0.94	0.96
	SMOTE-oversampled dataset4	0.90	0.95	0.85	0.96	0.90	0.96
	ADASYN-oversampled dataset	0.88	0.91	0.86	0.91	0.88	0.95
C5.0	Original data	0.94	0.96	0.47	0.98	0.97	0.82
	SMOTE-oversampled dataset1	0.92	0.96	0.90	0.93	0.95	0.94
	SMOTE-oversampled dataset4	0.94	0.96	0.90	0.97	0.93	0.93
	ADASYN-oversampled dataset	0.95	1	0.90	1	0.94	0.97
SVM	Original data	0.94	0.96	0.55	0.97	0.97	0.90
	SMOTE-oversampled dataset1	0.94	0.97	0.90	0.95	0.96	0.98
	SMOTE-oversampled dataset4	0.95	0.98	0.92	0.96	0.95	0.98
	ADASYN-oversampled dataset	0.94	0.95	0.82	0.93	0.88	0.97
Bayesian network	Original data	0.94	0.97	0.66	0.96	0.96	0.90
	SMOTE-oversampled dataset1	0.90	0.95	0.86	0.91	0.93	0.95
	SMOTE-oversampled dataset4	0.89	0.89	0.87	0.90	0.88	0.96
	ADASYN-oversampled dataset	0.88	0.89	0.86	0.89	0.88	0.94
CHAID tree	Original data	0.94	0.95	0.38	0.98	0.98	0.93
	SMOTE-oversampled dataset1	0.90	0.96	0.88	0.90	0.93	0.96
	SMOTE-oversampled dataset4	0.90	0.96	0.83	0.96	0.89	0.95
	ADASYN-oversampled dataset	0.88	0.89	0.86	0.90	0.88	0.95
Ensemble	Original data	0.94	0.96	0.48	0.98	0.97	0.94
	SMOTE-oversampled dataset1	0.92	0.96	0.88	0.93	0.95	0.98
	SMOTE-oversampled dataset4	0.95	0.97	0.84	0.97	0.95	0.98
	ADASYN-oversampled dataset	0.95	0.98	0.90	0.96	0.94	0.98

ISSN: 1472-6947