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

Table 6

Gait recognition accuracy for gait recognition mode C 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	37.96  10.39	30.65  6.43	33.30  4.62	37.53  10.16	32.38  3.62	31.35  5.74
IMF2	37.96  10.39	31.65  7.41	32.46  4.15	37.09  9.95	29.37  5.37	33.37  6.64
IMF3	37.96  10.39	39.06  8.80	40.21  7.41	35.76  9.01	42.92  9.60	41.02  8.74
IMF4	50.43  13.04	60.88  13.13	47.69  12.17	43.07  11.55	57.88  13.25	54.70  11.32
IMF5	38.65  10.57	72.87  16.95	54.08  15.26	49.54  16.93	51.17  12.55	62.38  15.30
IMF6	38.00  10.47	63.47  19.32	51.82  16.42	48.16  13.55	40.32  9.91	52.48  17.08
IMF7	42.30  18.02	50.23  18.64	40.93  12.69	42.34  17.73	40.47  8.71	40.40  14.34
IMF8	34.13  16.80	41.09  17.28	35.15  16.60	41.45  17.29	35.60  15.47	37.13  16.72