Decision Tree Ensembles to Predict Coronavirus Disease 2019 Infection: A Comparative Study

Table 5

The performance (AUROC) of various ensembles of decision trees for different sizes of ensembles.

Ensemble	Size
	20	50	100	200
Random forest	0.852	0.872	0.882	0.876
Bagging	0.859	0.858	0.856	0.857
XGBoost	0.863	0.864	0.867	0.868
AdaBoost	0.827	0.844	0.850	0.855
Ensembles for imbalanced datasets
Balanced random Forest (RUS)	0.871	0.879	0.889	0.893
SmoteBagging	0.857	0.846	0.854	0.859
RUSBagging	0.874	0.881	0.880	0.883
SmoteBoost	0.853	0.849	0.864	0.864
RUSBoost	0.7516	0.716	0.717	0.715

Bold numbers show the best performance.