An Efficient Automatic Gait Anomaly Detection Method Based on Semisupervised Clustering

Table 4

CCR/ACCR values for AGD-SSC, AGD-SSC-NBC, IF, and LOF in test case III.

Seq. No.	Metric	AGD-SSC; mean (st. dev.)	AGD-SSC-NBC; mean (st. dev.)	IF mean (st. dev.)	LOF mean (st. dev.)
Seq. 1	CCR	0.9444 (0.0000)	0.9444 (0.0000)	0.8821 (0.0089)	0.9074 (0.0000)
	ACCR	0.9167 (0.0000)	0.9167 (0.0000)	0.5222 (0.0369)	0.5833 (0.0000)
Seq. 2	CCR	0.9815 (0.0000)	0.9599 (0.0069)	0.9272 (0.0046)	0.8519 (0.0000)
	ACCR	1.0000 (0.0000)	0.9861 (0.0311)	0.7528 (0.015)	0.4167 (0.0000)
Seq. 3	CCR	0.9815 (0.0000)	0.9815 (0.0000)	0.8920 (0.0097)	0.8889 (0.0000)
	ACCR	0.9167 (0.0000)	0.9167 (0.0000)	0.5972 (0.0435)	0.5000 (0.0000)
Seq. 4	CCR	0.9284 (0.0063)	0.9265 (0.0033)	0.8988 (0.0149)	0.8889 (0.0000)
	ACCR	0.9167 (0.0000)	0.9167 (0.0000)	0.6167 (0.0593)	0.6667 (0.0000)
Seq. 5	CCR	0.9444 (0.0000)	0.8704 (0.0000)	0.9364 (0.0124)	0.8704 (0.0000)
	ACCR	0.9167 (0.0000)	0.9167 (0.0000)	0.7278 (0.0472)	0.4167 (0.0000)
Seq. 6	CCR	0.9815 (0.0000)	0.9630 (0.0000)	0.8796 (0.0104)	0.8889 (0.0000)
	ACCR	0.9167 (0.0000)	0.9167 (0.0000)	0.5417 (0.0469)	0.5000 (0.0000)
Seq. 7	CCR	0.9630 (0.0000)	0.9630 (0.0000)	0.8938 (0.0106)	0.9444 (0.0000)
	ACCR	1.000 (0.0000)	1.0000 (0.0000)	0.6111 (0.0393)	0.7500 (0.0000)
Seq. 8	CCR	0.9809 (0.0033)	0.9549 (0.0092)	0.9056 (0.0120)	0.8889 (0.0000)
	ACCR	09139 (0.0150)	0.9139 (0.0150)	0.6583 (0.0542)	0.5833 (0.0000)
Seq. 9	CCR	0.9815 (0.0000)	0.9444 (0.0000)	0.9019 (0.0137)	0.8889 (0.0000)
	ACCR	1.0000 (0.0000)	1.0000 (0.0000)	0.9000 (0.0333)	0.7500 (0.0000)
Seq. 10	CCR	0.9815 (0.0000)	0.9617 (0.0046)	0.9451 (0.0155)	0.9259 (0.0000)
	ACCR	1.0000 (0.0000)	1.0000 (0.0000)	0.7889 (0.0468)	0.6667 (0.0000)
Total	CCR	0.9669	0.9470	0.9062	0.8944
	ACCR	0.9497	0.9483	0.6717	0.5833