Ensemble Merit Merge Feature Selection for Enhanced Multinomial Classification in Alzheimer’s Dementia

Table 4

Results with the maximal feature subset obtained by divide and merge feature selection technique.

Classifier	Normal class			Dementia			Mild Cognitive Impairment
	Acc	Pre	Rec	Acc	Pre	Rec	Acc	Pre	Rec
J48	0.963	0.963	0.963	0.978	0.953	0.967	0.966	0.968	0.987
J48-TT	0.983	0.972	0.977	0.986	0.973	0.963	0.977	0.976	0.977
J48-CV	0.965	0.954	0.945	0.966	0.956	0.955	0.964	0.954	0.945

TT: training and testing; CV: cross validation.