An Infrastructure to Mine Molecular Descriptors for Ligand Selection on Virtual Screening

Table 3

Evaluation metric results of the second set of data mining experiments for generating rules about the molecular descriptors. Columns 1 and 2 are the definition of the decision tree experiment characteristics. Columns 3–7 correspond to the resulted metrics for each performed experiment: accuracy, size, RMSE, MAE, and 𝐹-measure, respectively.

Classes	Method	Accuracy	Size	RMSE	MAE	FM
2	1	84.15	9	0.35	0.22	0.84
	2	84.39	7	0.34	0.20	0.84
3	1	64.88	9	0.39	0.28	0.65
	2	77.81	13	0.33	0.20	0.78
4	1	58.78	17	0.36	0.25	0.59
	2	68.29	13	0.34	0.22	0.67