Skip to main content

Table 2 Average AUC of 10 different feature selection algorithms on 4 different microarray datasets

From: Multi-task feature selection in microarray data by binary integer programming

  

Colon

Lung

DLBCL

Myeloma

Average

m = 20

PC

.775 ± .159

.657 ± .184

.945 ± .051

.689 ± .094

.767

 

ChiSquare

.763 ± .189

.573 ± .146

.945 ± .043

.639 ± .121

.730

 

GINI

.760 ± .217

.590 ± .170

.948 ± .054

.653 ± .096

.738

 

InfoGain

.758 ± .197

.546 ± .160

.948 ± .054

.639 ± .111

.723

 

KW

.735 ± .145

.548 ± .165

.858 ± .099

.582 ± .112

.681

 

Relief

.775 ± .149

.685 ± .195

.949 ± .043

.671 ± .104

.770

 

mRMR

.785 ± .163

.556 ± .164

.938 ± .074

.649 ± .126

.732

 

SASMIF

.710 ± .168

.560 ± .145

.931 ± .052

.612 ± .076

.703

 

QSFS

.793 ± .129

.579 ± .186

.942 ± .043

.737 ± .062

.763

 

ST-BIP

.815 ± .153

.612 ± .108

.953 ± .054

.701 ± .048

.770

m = 50

PC

.763 ± .170

.648 ± .184

.958 ± .025

.709 ± .071

.770

 

ChiSquare

.740 ± .189

.600 ± .173

.965 ± .035

.676 ± .076

.745

 

GINI

.742 ± .183

.586 ± .167

.966 ± .034

.666 ± .096

.740

 

InfoGain

.755 ± .179

.595 ± .170

.963 ± .026

.682 ± .085

.749

 

KW

.755 ± .187

.574 ± .163

.858 ± .128

.606 ± .072

.698

 

Relief

.785 ± .145

.661±.194

.966 ± .027

.677 ± .082

.772

 

mRMR

.748 ± .182

.651 ± .219

.948 ± .067

.695 ± .093

.761

 

SASMIF

.663 ± .206

.563 ± .130

.943 ± .043

.636 ± .004

.701

 

QSFS

.695 ± .208

.608 ± .054

.961 ± .031

.714±.080

.745

 

ST-BIP

.828 ± .082

.600 ± .124

.969 ± .034

.710 ± .110

.777

m = 100

PC

.753 ± .176

.607 ± .122

.963 ± .025

.708 ± .062

.758

 

ChiSquare

.745 ± .184

.631 ± .164

.966 ± .024

.688 ± .063

.758

 

GINI

.748 ± .186

.594 ± .202

.965 ± .026

.698 ± .079

.751

 

InfoGain

.750 ± .180

.631 ± .164

.967 ± .022

.690 ± .062

.760

 

KW

.727 ± .188

.570 ± .206

.879 ± .113

.624 ± .071

.700

 

Relief

.773 ± .177

.631 ± .176

.958 ± .042

.708 ± .066

.768

 

mRMR

.758 ± .169

.608 ± .169

.966 ± .035

.690 ± .075

.756

 

SASMIF

.785 ± .131

.611 ± .213

.950 ± .035

.647 ± .072

.748

 

QSFS

.777 ± .173

.636±.113

.965 ± .025

.710 ± .073

.772

 

ST-BIP

.833 ± .078

.627 ± .180

.975 ± .033

.735 ± .086

.793

m = 200

PC

.760 ± .164

.632 ± .120

.973 ± .018

.704 ± .059

.767

 

ChiSquare

.750 ± .165

.611 ± .198

.973 ± .030

.673 ± .072

.752

 

GINI

.753 ± .165

.617 ± .199

.974 ± .019

.690 ± .064

.759

 

InfoGain

.755 ± .165

.611 ± .198

.977 ± .017

.673 ± .072

.754

 

KW

.735 ± .219

.571 ± .199

.878 ± .145

.637 ± .036

.705

 

Relief

.758 ± .162

.621 ± .157

.979 ± .025

.721±.076

.770

 

mRMR

.755 ± .155

.585 ± .169

.974 ± .027

.668 ± .068

.746

 

SASMIF

.820 ± .011

.590 ± .124

.954 ± .221

.644 ± .045

.752

 

QSFS

.765 ± .171

.664 ±.187

.974 ± .025

.687 ± .052

.773

 

ST-BIP

.833 ± .080

.634 ± .156

.984 ± .020

.706 ± .106

.789

m = 1000

PC

.740 ± .172

.633 ± .193

.979 ± .018

.700 ± .049

.763

 

ChiSquare

.743 ± .174

.606 ± .121

.974 ± .028

.676 ± .060

.750

 

GINI

.735 ± .176

.645 ±.152

.974 ± .027

.679 ± .056

.758

 

InfoGain

.743 ± .174

.606 ± .121

.974 ± .028

.676 ± .060

.750

 

KW

.722 ± .198

.568 ± .184

.941 ± .051

.652 ± .037

.721

 

Relief

.728 ± .173

.623 ± .150

.980 ± .019

.698 ± .051

.757

 

mRMR

.743 ± .174

.606 ± .121

.976 ± .025

.677 ± .060

.751

 

SASMIF

.763 ± .149

.587 ± .176

.952 ± .038

.669 ± .054

.743

 

QSFS

.745 ± .175

.624 ± .163

.980 ± .017

.690 ± .047

.760

 

ST-BIP

.828 ± .063

.625 ± .192

.981 ± .020

.722 ± .078

.789