Table 5.

Number of instances where the winning classifier trained on synthetic data matches the winning classifier trained on real data across 19 datasets.

Synthetic dataset	Winning classifier matches for real versus synthetic
	5 classifiers	4 classifiers (DTa removed)	3 classifiers (DT and RFb removed)
CARTc	5/19 (26.3)	10/19 (52.6)	14/19 (73.7)
Parametric	5/19 (26.3)	10/19 (52.6)	10/19 (52.6)
Bayesian network	4/19 (21.1)	10/19 (52.6)	13/19 (68.4)
All	14/57	30/57	37/57

^aDT: decision tree.

^bRF: random forest.

^cCART: classification and regression trees.