TABLE 2

Summary of results using all descriptors in training and test sets

Set	Method	Predicted Compounds							Correctly Predicted	MCC	5-Fold CV
Set	Method			TP	FP	TN	FN		Correctly Predicted	MCC	5-Fold CV
		%
Training set	SVM^D	80	20	85	14	83	0.66	71
Training set	SVM^E	100	0	100	0	100	1.00	70
Training set	RF	100	0	100	0	100	1.00	73
Training set	kNN	84	16	72	28	78	0.55	68
Training set	C4.5/J48	96	4	97	3	97	0.93	67
Test set	SVM^D	80	20	61	39	72	0.42
Test set	SVM^E	77	23	67	33	73	0.45
Test set	kNN	79	21	65	35	73	0.44
Test set	C4.5/J48	71	29	72	28	71	0.43

TP, number of true positives; FP, number of false positives; TN, number of true negatives; FN, number of false negatives; CV, cross-validation; SVM^D, support vector machine, linear model; SVM^E, support vector machine, nonlinear model; RF, random forest method; kNN, kappa nearest neighbors method; C4.5/J48, decision tree method.