Skip to main content

Table 10 Summary of results—subject 3 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

82.29

18.04

195

0.78

SVM-Corr

80.21

22.27

260

0.87

SVM-RFE

85.42

13.93

157

0.66

LASSO (\(\varTheta\) = 0.6)

68.75

14.60

75

0.60

LASSO (\(\varTheta\) = 0.7)

71.88

19.31

80

0.57

LASSO (\(\varTheta\) = 0.8)

64.58

18.34

79

0.58

LASSO (\(\varTheta\) = 0.9)

60.42

19.09

70

0.57

ENet (\(\varTheta\) = 0.6)

40.63

17.78

242

0.71

ENet (\(\varTheta\) = 0.7)

61.46

18.04

276

0.67

ENet (\(\varTheta\) = 0.8)

62.50

15.08

263

0.64

ENet (\(\varTheta\) = 0.9)

62.50

10.66

231

0.62