A Methylation Diagnostic Model Based on Random Forests and Neural Networks for Asthma Identification

Table 2

ROC validation results of three machine learning models (SVM, CART, and XGBoost).

Methods	GSE40576			GSE137716
	AUC	Specificity	Sensitivity	AUC	Specificity	Sensitivity
SVM	0.825	0.845	0.804	0.938	1.000	0.875
CART	0.773	0.856	0.691	0.818	0.824	0.812
XGBoost	0.619	0.619	0.619	0.881	0.824	0.938

CART: Classification and Regression Trees; SVM: support vector machines; XGBoost: eXtreme Gradient Boosting; AUC: Area Under Curve.