Skip to main content

Table 12 Summary of results—subject 5 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

64.77

30.53

142

0.62

SVM-Corr

68.18

29.24

255

0.77

SVM-RFE

65.91

29.63

237

0.72

LASSO (\(\varTheta\) = 0.6)

51.14

24.01

24

0.58

LASSO (\(\varTheta\) = 0.7)

46.59

21.72

23

0.57

LASSO (\(\varTheta\) = 0.8)

39.77

22.23

21

0.43

LASSO (\(\varTheta\) = 0.9)

15.91

9.83

22

0.12

ENet (\(\varTheta\) = 0.6)

45.45

21.12

66

0.54

ENet (\(\varTheta\) = 0.7)

39.77

27.28

67

0.52

ENet (\(\varTheta\) = 0.8)

46.59

23.78

65

0.53

ENet (\(\varTheta\) = 0.9)

48.86

27.07

76

0.59