Predicting Metabolic Syndrome Using the Random Forest Method

Table 4

Summary of prediction performance for MS classification using Random Forest from 20 independent runs.

Prediction performance	Internal test set (10-fold CV)				External test set
	Acc	Sens	Spec	MCC	Acc	Sens	Spec	MCC
Mean	97.88	94.54	98.91	0.94	98.12	94.80	99.15	0.95
SD	0.18	0.65	0.12	0.00	0.45	1.49	0.45	0.01

10-fold CV: 10-fold cross-validation, Acc: accuracy, Sens: sensitivity, Spec: specificity, and MCC: Matthews correlation coefficient.