Skip to main content

Table 5 Molecular subtype classification accuracy of Breast Cancer for the 1046 features

From: Automatic discovery of 100-miRNA signature for cancer classification using ensemble feature selection

 

Normal

LumA

LumB

TNBC

Her2

Global

#Samples

33

399

139

135

58

764

Gradient Boosting

0.1818

0.9348

0.5396

0.9333

0.5172

0.7987

Random Forest

0.0606

0.9724

0.4532

0.9630

0.0345

0.7657

Logistic Regression

0.1212

0.8747

0.5540

0.9259

0.4483

0.7606

Passive Aggressive

0.1515

0.8622

0.5612

0.9111

0.4483

0.7539

SGD

0.3030

0.9073

0.4604

0.9556

0.4655

0.7752

SVC

0.2727

0.8797

0.5252

0.9185

0.5345

0.7697

Ridge

0.1515

0.7293

0.4317

0.3704

0.2759

0.5524

Bagging

0.3333

0.9298

0.5108

0.9704

0.4310

0.7973

Average

0.1970

0.8863

0.5045

0.8685

0.3944

0.7467