Table 3.

Model selection results for the seven machine learning methods

	PPV	NPV	AUC	F1-score
All features
RF	0.030	0.998	0.88	0.89
ANN	0.028	0.998	0.86	0.86
XGB	0.026	0.998	0.88	0.87
ADA	0.028	0.998	0.89	0.87
LR	0.026	0.999	0.88	0.87
kNN	0.025	0.998	0.86	0.83
SVM	0.024	0.999	0.89	0.86

Bold italics indicate the best model for a given parameter

PPV positive predictive value, NPV negative predictive value, AUC area under the curve of the ROC analysis, ACC accuracy