Skip to main content

Table 8 Summary of results—subject 1 in Haxby dataset

From: Exploring stability-based voxel selection methods in MVPA using cognitive neuroimaging data: a comprehensive study

Method

Mean accuracy (%)

STD (%)

Average number of selected features

Stability

SVM-MI

90.63

12.07

122

0.70

SVM-Corr

84.38

16.96

338

0.58

SVM-RFE

84.38

16.96

219

0.49

LASSO (\(\varTheta\) = 0.6)

79.17

14.43

88

0.68

LASSO (\(\varTheta\) = 0.7)

77.08

13.93

75

0.67

LASSO (\(\varTheta\) = 0.8)

76.04

11.25

87

0.71

LASSO (\(\varTheta\) = 0.9)

71.88

16.10

95

0.61

ENet (\(\varTheta\) = 0.6)

38.54

26.36

255

0.71

ENet (\(\varTheta\) = 0.7)

59.38

20.03

235

0.70

ENet (\(\varTheta\) = 0.8)

62.50

21.98

255

0.67

ENet (\(\varTheta\) = 0.9)

80.21

11.25

232

0.67