Table 5.

Accuracy (ACC), precision (PPV, Positive Predictive Value) and sensitivity (TPR, True Positive Rate) for each functional fitness movement, related to the cubic kernel SVM algorithm with a 600 ms length windows and 10% overlap.

	ACC		PPV		TPR
	Stage 1	Stage 2	Stage 1	Stage 2	Stage 1	Stage 2
C&J	99.4	97.3	$99.1$	$89.3$	91.2	82.2
AS	99.1	98.7	$93.0$	93.0	$81.1$	$79.3$
BJ	99.5	$99.2$	94.8	93.5	87.6	85.7
BP	$99.7$	98.8	96.5	92.7	96.5	86.2
TRANS	$97.9$	$94.3$	98.0	$94.9$	$99.4$	$97.8$

Data from all the five IMUs available were used as input. Classification performance is reported from the two stages of validation as %, and results are the average across multiple iterations. Stage 1= 5-fold Cross-Validation—Training Dataset; Stage 2 = “leave one subject out” (LOSO) on test dataset. Green bold numbers = best performance; red bold numbers = worst performance.