Table B2.

Confusion matrix and performance values for the ANN-based classification system.

Test/Real	Walk	Jump	Immobile	Run	Up	Down	Cycle	Drive	Instances	TP	TN	FP	FN	Accuracy	Recall	Specificity	Precision	F1-Measure
Walk	6930	46	41	117	107	79	86	121	7527	6930	51,807	597	619	97.97%	91.80%	98.86%	92.07%	91.93%
Jump	76	7124	16	187	13	8	46	25	7495	7124	51,892	371	566	98.44%	92.64%	99.29%	95.05%	93.,83%
Immobile	47	9	6970	4	63	46	13	246	7398	6970	52,307	428	248	98.87%	96.56%	99.19%	94.21%	95.37%
Run	44	221	3	7060	39	43	59	28	7497	7060	52,016	437	440	98.54%	94.13%	99.17%	94.17%	94.15%
Up	126	89	17	25	7091	256	19	16	7639	7091	51,587	548	727	97.87%	90.70%	98.95%	92.83%	91.75%
Down	103	4	7	27	415	7103	16	21	7696	7103	51,754	593	503	98.17%	93.39%	98.87%	92.29%	92.84%
Cycle	36	32	43	35	75	56	6928	98	7303	6928	52,386	375	264	98.93%	96.33%	99.29%	94.87%	95.59%
Drive	187	165	121	45	15	15	25	6825	7398	6825	52,000	573	555	98.12%	92.48%	98.91%	92.25%	92.37%
Accumulated	7549	7690	7218	7500	7818	7606	7192	7380	59,953	56,031	415,749	3922	3922	98.36%	93.50%	99.07%	93.47%	93.48%