The Ensemble Machine Learning-Based Classification of Motor Imagery Tasks in Brain-Computer Interface

Table 1

Summary of the classification performance measures.

Subj.	Classifier	Accuracy		F-measure		AUC
AA		RSM	RoF	RSM	RoF	RSM	RoF
	SVM	89.64	90.60	0.896	0.906	0.958	0.929
	k-NN	93.45	96.67	0.935	0.967	0.988	0.985
	RF	82.02	84.76	0.820	0.848	0.900	0.928
	C4.5	67.50	74.76	0.675	0.748	0.764	0.834
	REP tree	69.88	67.50	0.699	0.675	0.765	0.746
	RT	70.83	74.29	0.706	0.742	0.799	0.827
AL	SVM	86.43	84.88	0.864	0.849	0.923	0.879
	k-NN	92.26	94.05	0.923	0.940	0.970	0.968
	RF	75.00	77.26	0.750	0.773	0.830	0.852
	C4.5	71.07	73.69	0.711	0.737	0.786	0.812
	REP tree	68.93	68.69	0.689	0.687	0.751	0.756
	RT	70.95	70.60	0.708	0.703	0.778	0.786
AV	SVM	87.02	84.40	0.870	0.844	0.928	0.886
	k-NN	88.93	89.64	0.889	0.896	0.955	0.939
	RF	73.69	76.90	0.737	0.769	0.818	0.857
	C4.5	61.90	71.31	0.619	0.713	0.677	0.767
	REP tree	62.38	66.19	0.624	0.662	0.671	0.721
	RT	65.00	65.95	0.644	0.654	0.691	0.738
AW	SVM	86.90	85.00	0.869	0.850	0.926	0.876
	k-NN	94.64	96.43	0.946	0.964	0.983	0.976
	RF	77.02	78.81	0.770	0.788	0.844	0.867
	C4.5	67.62	71.19	0.676	0.712	0.727	0.788
	REP tree	63.45	68.33	0.634	0.683	0.683	0.742
	RT	68.69	70.24	0.685	0.700	0.735	0.780
AY	SVM	92.02	91.90	0.920	0.919	0.984	0.942
	k-NN	92.38	90.71	0.924	0.907	0.977	0.952
	RF	98.69	94.29	0.987	0.943	0.999	0.991
	C4.5	98.45	97.14	0.985	0.971	0.999	0.995
	REP tree	95.60	93.10	0.956	0.931	0.988	0.980
	RT	97.02	93.10	0.970	0.931	0.996	0.985
ALL	SVM	89.98	90.12	0.900	0.901	0.962	0.932
	k-NN	93.55	94.83	0.935	0.948	0.984	0.974
	RF	80.36	81.83	0.804	0.818	0.889	0.901
	C4.5	68.57	73.00	0.686	0.730	0.769	0.815
	REP tree	68.60	69.02	0.686	0.690	0.765	0.760
	RT	72.57	72.40	0.724	0.722	0.807	0.809