Motor Imagery EEG Classification Based on Decision Tree Framework and Riemannian Geometry

Table 1

Ten-folder cross-validation classification accuracy (%) for FGMDRM with OVR scheme applied on BCI competition dataset 2A.

Subject	A01	A02	A03	A04	A05	A06	A07	A08	A09	Mean	Std
L/RE	84.40	67.39	93.05	77.64	66.96	74.34	86.33	93.83	94.09	82.00	10.90
R/RE	89.26	75.33	95.16	78.56	63.92	71.81	81.18	93.39	82.92	81.28	10.24
F/RE	77.80	83.44	89.20	80.88	71.56	78.22	88.89	79.18	84.4	81.51	5.65
T/RE	88.21	68.76	90.59	79.95	71.85	76.72	90.64	94.43	94.38	83.95	9.82