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

Table 4

The performance of Standard decision tree ensembles with RUS undersampling method for different ratios of minority and majority class points. AUROC and AUPRC performance measures are used in the experiments.

Ensemble	Original		Ratio = 1		Ratio = 0.75		Ratio = 0.50
	AUROC	AUPRC	AUROC	AUPRC	AUROC	AUPRC	AUROC	AUPRC
Single decision tree	0.659	0.307	0.731	0.231	0.733	0.26	0.745	0.291
Random forest	0.872	0.499	0.852	0.454	0.859	0.476	0.855	0.454
Bagging	0.858	0.506	0.839	0.422	0.848	0.445	0.835	0.442
XGBoost	0.863	0.505	0.845	0.475	0.846	0.510	0.873	0.554
AdaBoost	0.844	0.410	0.826	0.389	0.830	0.401	0.849	0.425

Bold numbers show the best results.