Table 8.

Statistical evaluation metrics of the classifiers for feature sets S1 and S2

Classifier	Statistical parameters of the classification for feature set S1								Class
	TP	FP	Prec.	Rec.	F-S.	ROC	Sen.	Spec.
Random Forest	0.949	0.017	0.974	0.949	0.961	0.998	0.949	0.983	Early
	0.973	0.017	0.973	0.973	0.973	0.996	0.973	0.983	Variable
	1.0	0.013	0.955	1.0	0.977	1.0	1.0	0.987	Late
MLP	0.974	0.017	0.974	0.974	0.974	0.999	0.974	0.983	Early
	0.973	0.013	0.973	0.973	0.973	0.993	0.973	0.987	Variable
	1.0	0	1.0	1.0	1.0	1.0	0.999	0.998	Late
Naïve Bayes	0.872	0.017	0.971	0.872	0.919	0.989	0.872	0.983	Early
	0.919	0.133	0.810	0.919	0.861	0.938	0.919	0.867	Variable
	0.857	0.026	0.900	0.857	0.878	0.977	0.857	0.974	Late
Simple Logistics	0.949	0.017	0.974	0.949	0.961	0.987	0.949	0.983	Early
	0.973	0.050	0.923	0.973	0.947	0.986	0.973	0.950	Variable
	0.905	0.013	0.950	0.905	0.927	0.986	0.905	0.987	Late

Classifier	Statistical parameters of the classification for feature set S2								Class
	TP	FP	Prec.	Rec.	F-S.	ROC	Sen.	Spec.
Random Forest	0.625	0.211	0.676	0.625	0.649	0.789	0.625	0.788	Early
	0.740	0.468	0.627	0.740	0.679	0.726	0.740	0.532	Variable
	0	0.011	0	0	0	0.464			Late
MLP	0.625	0.316	0.581	0.625	0.602	0.639	0.625	0.680	Early
	0.620	0.383	0.633	0.620	0.626	0.644	0.620	0.617	Variable
	0	0.056	0	0	0	0.483			Late
Naïve Bayes	0.450	0.193	0.621	0.450	0.522	0.646	0.48	0.785	Early
	0.660	0.511	0.579	0.660	0.617	0.591	0.66	0.49	Variable
	0	0.122	0	0	0	0.417			Late
Simple Logistics	0.600	0.263	0.615	0.600	0.608	0.668	0.66	0.737	Early
	0.600	0.404	0.612	0.600	0.606	0.598	0.61	0.61	Variable
	0	1.0	0	0	0	0.450			Late

TP: True Positive, FP: False Positive, Prec.: Precision, Rec.: Recall, F-S.: F-Score, ROC: Receiver Operating Characteristic Curve, Sen.: Sensitivity, Spec.: Specificity.