A Fused Multidimensional EEG Classification Method Based on an Extreme Tree Feature Selection

Table 3

Classification accuracy for datasets III BCI 2003 and 2b BCI 2008 using different feature extraction methods.

Dataset	Public datasets
	III BCI 2003			2b BCI 2008
	aa	bb	cc	dd	ee	ff
AR + SVM	0.622	0.416	0.397	0.649	0.551	0.596
CSP + SVM	0.420	0.493	0.580	0.786	0.541	0.622
DWT + SVM	0.451	0.573	0.668	0.742	0.657	0.585
DWT + CSP + SVM	0.693	0.659	0.775	0.808	0.617	0.658
AR + DWT + SVM	0.636	0.725	0.667	0.792	0.758	0.700
AR + CSP + SVM	0.650	0.783	0.633	0.808	0.683	0.675
AR + CSP + DWT + SVM	0.650	0.725	0.742	0.817	0.733	0.675