Skip to main content

Table 2 Results of 16 datasets (16 binary classification tasks) using different data scaling algorithms and classification models

From: A robust data scaling algorithm to improve classification accuracies in biomedical data

dataset

Method

None

Minmax

Zscore

GL

GSE27899IL

LR

NA ± NA

NA ± NA

NA ± NA

NA ± NA

 

SVM

0.768 ± 0.104

0.814 ± 0.084

0.814 ± 0.074

0.824 ± 0.058

Prostate Cancer

LR

0.464 ± 0.000

0.749 ± 0.130

0.689 ± 0.156

0.761 ± 0.108

 

SVM

0.573 ± 0.198

0.725 ± 0.232

0.713 ± 0.244

0.822 ± 0.194

Colon Cancer

LR

0.500 ± 0.000

0.895 ± 0.092

0.892 ± 0.082

0.962 ± 0.046

 

SVM

0.670 ± 0.184

0.940 ± 0.058

0.937 ± 0.050

0.981 ± 0.020

Lung Cancer

LR

0.450 ± 0.000

0.839 ± 0.096

0.834 ± 0.108

0.890 ± 0.050

 

SVM

0.397 ± 0.274

0.716 ± 0.136

0.710 ± 0.152

0.774 ± 0.182

Breast Cancer

LR

0.324 ± 0.000

0.809 ± 0.038

0.821 ± 0.020

0.819 ± 0.022

 

SVM

0.708 ± 0.158

0.793 ± 0.052

0.795 ± 0.042

0.812 ± 0.038

Leukemia

LR

0.500 ± 0.000

0.988 ± 0.014

0.990 ± 0.006

1.000 ± 0.000

 

SVM

0.935 ± 0.034

0.992 ± 0.010

0.991 ± 0.008

1.000 ± 0.000

GSE29490

LR

NA ± NA

NA ± NA

NA ± NA

NA ± NA

 

SVM

0.983 ± 0.012

0.984 ± 0.034

0.985 ± 0.034

0.994 ± 0.004

GSE25869

LR

NA ± NA

NA ± NA

NA ± NA

NA ± NA

 

SVM

0.935 ± 0.024

0.937 ± 0.020

0.938 ± 0.016

0.943 ± 0.014

Breast tissue

LR

0.520 ± 0.006

0.961 ± 0.032

0.961 ± 0.044

0.940 ± 0.054

 

SVM

0.713 ± 0.108

0.968 ± 0.006

0.970 ± 0.014

0.972 ± 0.010

LSVT

LR

0.500 ± 0.000

0.875 ± 0.008

0.846 ± 0.022

0.921 ± 0.012

 

SVM

0.500 ± 0.000

0.879 ± 0.012

0.863 ± 0.014

0.919 ± 0.020

DLBCL

LR

0.601 ± 0.038

0.608 ± 0.038

0.610 ± 0.048

0.660 ± 0.062

 

SVM

0.616 ± 0.050

0.622 ± 0.052

0.619 ± 0.052

0.654 ± 0.054

Myeloma

LR

0.500 ± 0.000

0.729 ± 0.044

0.739 ± 0.072

0.746 ± 0.038

 

SVM

0.573 ± 0.098

0.748 ± 0.052

0.747 ± 0.054

0.750 ± 0.054

Parkinsons

LR

0.875 ± 0.012

0.896 ± 0.054

0.893 ± 0.058

0.906 ± 0.048

 

SVM

0.882 ± 0.010

0.875 ± 0.010

0.885 ± 0.024

0.891 ± 0.018

Wdbc

LR

0.942 ± 0.002

0.982 ± 0.004

0.978 ± 0.006

0.993 ± 0.004

 

SVM

0.990 ± 0.002

0.994 ± 0.002

0.993 ± 0.004

0.995 ± 0.000

Indian Liver

LR

0.680 ± 0.002

0.743 ± 0.008

0.742 ± 0.008

0.746 ± 0.010

 

SVM

0.636 ± 0.068

0.696 ± 0.008

0.692 ± 0.034

0.695 ± 0.008

Pima Indians Diabetes

LR

0.604 ± 0.004

0.827 ± 0.004

0.827 ± 0.004

0.834 ± 0.006

 

SVM

0.826 ± 0.004

0.828 ± 0.006

0.828 ± 0.006

0.834 ± 0.006

  1. The performances are measured by the average Area Under the ROC in 5-fold cross-validations. The means and 95 % confidence intervals are included. Column names: None - no data scaling; Minmax - Min-max algorithm; Z-score - Z-score algorithm; GL - GL algorithm. Best performances are emphasized in bold