Data

Model

Normalization

None

Z-standard

Min-Max

MaxAbs (−1, 1)

Quantile Transform

Quantile Normalize

Ranked - Binary Target

Logistic

Total Loss

0.444

0.505

0.506

0.506

0.506

0.444

Bias

0.369

0.343

0.344

0.344

0.343

0.369

Variance

0.075

0.163

0.162

0.240

0.163

0.075

Noise

0.000

0.000

0.000

0.078

0.000

0.000

Variance-Bias Ratio

0.202

0.474

0.473

0.698

0.476

0.202

Percent Change from Raw

~

113.941

114.128

114.127

114.080

100.000

Decision Tree

Total Loss

0.489

0.494

0.494

0.494

0.494

0.493

Bias

0.292

0.246

0.246

0.246

0.246

0.291

Variance

0.197

0.248

0.248

0.454

0.248

0.202

Noise

0.000

0.000

0.000

0.206

0.000

0.000

Variance-Bias Ratio

0.672

1.009

1.006

1.843

1.006

0.692

Percent Change from Raw

~

100.905

101.041

101.041

101.041

100.767

Random Forest

Total Loss

0.445

0.488

0.487

0.487

0.487

0.445

Bias

0.354

0.248

0.247

0.247

0.247

0.354

Variance

0.091

0.240

0.240

0.399

0.240

0.091

Noise

0.000

0.000

0.000

0.159

0.000

0.000

Variance-Bias Ratio

0.257

0.965

0.969

1.613

0.969

0.257

Percent Change from Raw

~

109.658

109.524

109.524

109.524

100.000

SVM

Total Loss

0.437

0.441

0.460

0.460

0.459

0.437

Bias

0.437

0.407

0.308

0.309

0.313

0.437

Variance

0.000

0.034

0.152

0.187

0.146

0.000

Noise

0.000

0.000

0.000

0.036

0.000

0.000

Variance-Bias Ratio

0.000

0.083

0.493

0.606

0.468

0.000

Percent Change from Raw

~

101.023

105.432

105.403

105.171

100.000

Gradient Boosting

Total Loss

0.465

0.521

0.521

0.521

0.521

0.465

Bias

0.303

0.298

0.299

0.299

0.299

0.303

Variance

0.162

0.222

0.222

0.334

0.222

0.162

Noise

0.000

0.000

0.000

0.112

0.000

0.000

Variance-Bias Ratio

0.535

0.745

0.745

1.119

0.745

0.535

Percent Change from Raw

~

112.059

112.123

112.123

112.123

100.000