Table 3.
Results of 16 datasets (16 binary classification tasks) using different data scaling algorithms and classification models
| dataset | Method | None | Minmax | Zscore | GL |
|---|---|---|---|---|---|
| GSE27899IL | LR | NA ± NA | NA ± NA | NA ± NA | NA ± NA |
| SVM | 0.770 ± 0.054 | 0.780 ± 0.134 | 0.780 ± 0.134 | 0.780 ± 0.134 | |
| Prostate Cancer | LR | 0.609 ± 0.000 | 0.757 ± 0.132 | 0.722 ± 0.078 | 0.765 ± 0.100 |
| SVM | 0.635 ± 0.078 | 0.748 ± 0.156 | 0.748 ± 0.156 | 0.835 ± 0.114 | |
| Colon Cancer | LR | 0.577 ± 0.000 | 0.877 ± 0.064 | 0.877 ± 0.064 | 0.923 ± 0.054 |
| SVM | 0.677 ± 0.178 | 0.900 ± 0.042 | 0.915 ± 0.034 | 0.946 ± 0.042 | |
| Lung Cancer | LR | 0.741 ± 0.000 | 0.859 ± 0.062 | 0.852 ± 0.052 | 0.896 ± 0.062 |
| SVM | 0.778 ± 0.052 | 0.859 ± 0.034 | 0.859 ± 0.034 | 0.867 ± 0.040 | |
| Breast Cancer | LR | 0.773 ± 0.000 | 0.918 ± 0.040 | 0.955 ± 0.000 | 0.955 ± 0.000 |
| SVM | 0.827 ± 0.040 | 0.909 ± 0.000 | 0.909 ± 0.000 | 0.936 ± 0.050 | |
| Leukemia | LR | 0.710 ± 0.000 | 0.956 ± 0.030 | 0.965 ± 0.030 | 1.000 ± 0.000 |
| SVM | 0.939 ± 0.030 | 0.965 ± 0.030 | 0.965 ± 0.030 | 1.000 ± 0.000 | |
| GSE29490 | LR | NA ± NA | NA ± NA | NA ± NA | NA ± NA |
| SVM | 0.942 ± 0.034 | 0.954 ± 0.034 | 0.958 ± 0.034 | 0.979 ± 0.000 | |
| GSE25869 | LR | NA ± NA | NA ± NA | NA ± NA | NA ± NA |
| SVM | 0.891 ± 0.038 | 0.891 ± 0.044 | 0.894 ± 0.034 | 0.897 ± 0.034 | |
| Breast tissue | LR | 0.778 ± 0.016 | 0.930 ± 0.010 | 0.930 ± 0.016 | 0.927 ± 0.016 |
| SVM | 0.681 ± 0.220 | 0.932 ± 0.024 | 0.926 ± 0.020 | 0.942 ± 0.008 | |
| LSVT | LR | 0.500 ± 0.000 | 0.870 ± 0.012 | 0.824 ± 0.038 | 0.915 ± 0.002 |
| SVM | 0.500 ± 0.000 | 0.873 ± 0.036 | 0.858 ± 0.036 | 0.908 ± 0.006 | |
| DLBCL | LR | 0.567 ± 0.014 | 0.571 ± 0.014 | 0.579 ± 0.032 | 0.602 ± 0.074 |
| SVM | 0.594 ± 0.082 | 0.592 ± 0.064 | 0.585 ± 0.044 | 0.600 ± 0.100 | |
| Myeloma | LR | 0.792 ± 0.000 | 0.805 ± 0.020 | 0.804 ± 0.018 | 0.805 ± 0.026 |
| SVM | 0.794 ± 0.006 | 0.809 ± 0.014 | 0.807 ± 0.026 | 0.813 ± 0.020 | |
| Parkinsons | LR | 0.865 ± 0.006 | 0.894 ± 0.022 | 0.891 ± 0.016 | 0.868 ± 0.006 |
| SVM | 0.880 ± 0.016 | 0.884 ± 0.006 | 0.877 ± 0.020 | 0.868 ± 0.016 | |
| Wdbc | LR | 0.878 ± 0.002 | 0.965 ± 0.010 | 0.963 ± 0.012 | 0.971 ± 0.012 |
| SVM | 0.960 ± 0.010 | 0.979 ± 0.004 | 0.976 ± 0.002 | 0.980 ± 0.008 | |
| Indian Liver | LR | 0.716 ± 0.002 | 0.727 ± 0.014 | 0.733 ± 0.008 | 0.736 ± 0.006 |
| SVM | 0.719 ± 0.006 | 0.720 ± 0.014 | 0.718 ± 0.010 | 0.720 ± 0.008 | |
| Pima Indians Diabetes | LR | 0.490 ± 0.070 | 0.738 ± 0.010 | 0.738 ± 0.010 | 0.740 ± 0.012 |
| SVM | 0.734 ± 0.052 | 0.765 ± 0.008 | 0.753 ± 0.040 | 0.748 ± 0.034 |
The performances are measured by the average proportion of correct classification in 5-fold cross-validations. The means and 95 % confidence intervals are included. Column names: None - no data scaling; Minmax - Min-max algorithm; Zscore - Z-score algorithm; GL - GL algorithm. Best performances are emphasized in bold