BioMedical Engineering OnLine

Table 2 Classification accuracy for the real BCI dataset

From: L1 Norm based common spatial patterns decomposition for scalp EEG BCI

Subjects	Classification accuracy
Subjects	CSP	L1-SVD-CSP	TR-CSP
CXY	0.90	0.90	0.84
WZQ	0.66	0.69	0.66
FNX	0.66	0.67	0.67
GK	0.93	0.94	0.94
LPY	0.76	0.78	0.79
JSL	0.57	0.63	0.65
LB	0.70	0.79	0.70
MXY	0.61	0.62	0.61
SG	0.58	0.62	0.58
WCF	0.80	0.86	0.80
WH	0.59	0.62	0.59
XXC	0.93	0.96	0.95
XJP	0.98	0.99	0.98
WXY	0.74	0.75	0.71
Mean result	0.74 ± 0.14	0.77 ± 0.13‡ *	0.75 ± 0.13

The McNemar test is performed to investigate the recognition difference among the three spatial filters, values in bold denote the better result. ‘‡’ indicates the significance between CSP and L1-SVD-CSP (p < 0.05); ‘*’ indicates the significance between TR-CSP and L1-SVD-CSP (p < 0.05).

Back to article page

ISSN: 1475-925X

Contact us

Submission enquiries: journalsubmissions@springernature.com