Table 7. Performance comparison of data-4 (BCI Competition IV Dataset IIA).

Reference	Feature Extraction	Feature Selection	Classifier	Classification Accuracy									Mean	STD	Kappa									Mean	STD
				A01	A02	A03	A04	A05	A06	A07	A08	A09			A01	A02	A03	A04	A05	A06	A07	A08	A09
Uribe et al. (2019)	Correntropy	Wrapper	LDA or ELM												0.66	0.27	0.69	0.36	0.23	0.32	0.46	0.66	0.64	0.47	0.18
Rahman et al. (2020)	WPT	Rényi min-entropy	MLP-ANN	80	82	84	74	78	85	85	79	74	80.11	4.28
Gaur et al. (2018)	Covariance matrix	–	RG	91.49	60.56	94.16	76.72	58.52	68.52	78.57	97.01	93.85	79.93	14.9	0.86	0.24	0.70	0.68	0.36	0.34	0.66	0.75	0.82	0.60	0.22
Rodrigues et al. (2019)	STR	FS	least squares	60	33	67	45	33	33	35	70	67	49	15	0.46	0.13	0.56	0.26	0.11	0.11	0.16	0.60	0.56	0.33	0.20
She et al. (2019)			HSS-ELM	92.10	70.26	94.40	80.86	71.41	78.31	92.10	89.22	92.67	84.59	9.54										0.79
Zhang et al. (2019)	FBCSP		CNN-LSTM	89	69	92	82	84	68	95	90	92	84	9	0.87	0.59	0.90	0.76	0.82	0.66	0.95	0.89	0.89	0.81	0.12
She et al. (2020)			BGR-SSELM												0.79	0.46	0.81	0.59	0.38	0.43	0.81	0.82	0.81	0.65	0.18
Olivas-Padilla & Chacon-Murguia (2019)	DFBCSP		CNN	84.9	66.3	84.7	81.3	79.2	70.6	86.1	83.8	83.0	80.0	6.52	0.68	0.36	0.69	0.62	0.60	0.45	0.71	0.72	0.66	0.61	0.12
Liao et al. (2020)	Global spatial filters (FBCSP)		CNN	88.6	55.9	86.7	71.0	66.5	56.0	88.4	80.9	77.1	74.6	13.0	0.84	0.4	0.82	0.61	0.54	0.41	0.84	0.74	0.69	0.65	0.18
Proposed method				98.9	97.6	97.6	93.8	80.2	88.6	71.2	94.8	90.3	90.3	9.24	0.97	0.93	0.93	0.83	0.47	0.69	0.23	0.86	0.74	0.74	0.25