Predicting functional performance via classification of lower extremity strength in older adults with exergame-collected data

Table 2 Best classification results using a Logistic Model Tree

Algorithm: Logistic Model Tree, accuracy 91.358%	Correctly classified	Incorrectly classified	TP rate	FP rate	Precision	Recall	F	MCC	ROC area	PRC area
Not fit	29 (TP)	5 (FN)	0.853	0.043	0.935	0.853	0.892	0.823	0.940	0.946
Fit	45 (TN)	2 (FP)	0.957	0.147	0.900	0.957	0.928	0.823	0.940	0.930
Weighted average	74	7	0.914	0.103	0.915	0.914	0.913	0.823	0.940	0.936

TP true positive, FP false positive, F F-measure, MCC Matthews correlation coefficient, ROC receiver-operating characteristic curve, PRC precision-recall curve

ISSN: 1743-0003