Gait Recognition for Human-Exoskeleton System in Locomotion Based on Ensemble Empirical Mode Decomposition

Table 4

Gait recognition accuracy of gait recognition mode B based on the IMFs of the original signal for the 14 subjects who walked without exoskeletons (mean  SD%).

Component	SVM	Kmeans	Decision tree	Logistic regression	Naive Bayes	Random forest
IMF1	36.85  10.04	35.14  4.86	34.36  3.22	36.72  9.81	34.13  4.66	35.12  4.68
IMF2	36.88  10.09	35.94  4.75	35.99  3.83	37.04  9.81	32.89  5.35	36.67  5.22
IMF3	37.27  10.31	44.75  9.79	43.39  8.43	36.77  8.89	43.37  12.52	45.98  10.36
IMF4	63.27  14.92	71.30  13.18	58.90  10.17	48.55  12.15	69.78  12.30	67.57  11.58
IMF5	51.14  18.31	80.24  12.68	65.07  13.16	63.80  16.31	70.29  13.26	72.85  12.66
IMF6	39.60  10.32	69.88  17.20	61.95  17.14	64.91  13.19	55.16  14.47	68.83  14.67
IMF7	45.20  13.34	50.94  15.79	45.98  12.26	51.83  11.52	42.92  11.51	45.78  14.22
IMF8	37.52  10.29	48.20  20.72	43.78  13.13	48.74  14.82	40.31  9.48	43.88  12.93