Skip to main content

Table 4 Discrimination results from all methods

From: A novel meta-analysis based on data augmentation and elastic data shared lasso regularization for gene expression

Method

Training error

Test-1 accuracy (%)

Test-2 accuracy (%)

No. of selected genes

Convergence time (s)

Lasso

5.09% (0.002)

89.35

84.12

176

11.09

Elastic Net

4.68% (0.001)

91.35

90.12

337

9.52

HLR

4.57% (0.005)

89.98

86.04

139

120.02

SGL

4.42% (0.002)

88.21

85.24

127

63.87

Meta-Lasso

3.68% (0.001)

91.21

90.24

53

70.12

DSL

3.55% (0.000)

92.72

90.88

46

33.87

DA-DSL-L2

2.23% (0.002)

94.98

92.16

59

428.65

  1. Standard deviation is shown in brackets