Skip to main content

Table 3 Comparative analysis of proposed gene selection algorithm UMVMO-select with other existing gene selection methods with respect to sample classification outputs

From: Multi-view feature selection for identifying gene markers: a diversified biological data driven approach

Algorithm

Sensitivity

Specificity

F-score

CA

Prostate

DLBCL

Child ALL

Prostate

DLBCL

CHild ALL

Prostate

DLBCL

Child ALL

Prostate

DLBCL

Child ALL

UMVMO-select

0.9038

0.948

0.8

0.92

0.842

0.833

0.9125

0.948

0.8

0.91176

0.922

0.8181

Acharya et al. [3]

0.8846

0.9137

0.72

0.92

0.7894

0.8166

0.9019

0.921

0.742

0.9019

0.8831

0.7727

Graph-MPSO [5]

0.8962

0.9111

0.752

0.9

0.9207

0.8233

0.9002

0.8428

0.7671

0.898

0.9184

0.7909

Graph-SingleObjective (Correlation)[5]

0.8221

0.6389

0.71

0.855

0.8966

0.8042

0.8382

0.639

0.7295

0.8382

0.8355

0.7614

Graph-SingleObjective (SNR)[5]

0.8701

0.8333

0.64

0.865

0.8707

0.8442

0.8704

0.7434

0.7079

0.8676

0.8618

0.7568

T-test [24]

0.7778

0.7284

0.4640

0.8244

0.9119

0.68

0.8336

0.7052

0.5184

0.8497

0.8486

0.5964

RankSum test[23]

0.8547

0.7654

0.4640

0.8375

0.8945

0.87

0.8522

0.7327

0.5506

0.8768

0.8621

0.6855

MRMR[25]

0.9176

0.8889

0.7486

0.8686

0.9163

0.8762

0.8970

0.8244

0.7896

0.8936

0.9098

0.7782

Acharya et al.[15]

0.865

0.8965

0.68

0.88

0.7368

0.8

0.8735

0.904

0.708

0.8725

0.8571

0.7454

Saha et al. [14]

0.8461

0.8793

0.64

0.86

0.6842

0.7833

0.8543

0.8869

0.6736

0.8529

0.8311

0.7181