Table 3 Performance of original algorithms before applying the proposed methods
Algo. | Train % | Test % | Over- Fitting Diff. % | Pre | Rec | F1- Score | NO.F | F-Time (sec) | C-Time (sec) | AUC | Var. | ACC % |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
RNA gene dataset | ||||||||||||
 RFE | 100.000 | 99.800 | 0.200 | 0.999 | 0.998 | 0.998 | 10,265 | 190,000 | 60.000 | 1.000 | 0.00002 | 99.800 |
 RFS | 100.000 | 99.800 | 0.200 | 0.999 | 0.998 | 0.998 | 374.000 | 13.015 | 0.275 | 1.000 | 0.00002 | 99.800 |
DNA CNV dataset | ||||||||||||
 RFE | 97.500 | 87.000 | 10.500 | 0.741 | 0.706 | 0.709 | 8190 | 182,295 | 40.000 | 0.960 | 0.023125 | 87.000 |
 RFS | 89.803 | 84.054 | 5.749 | 0.819 | 0.764 | 0.775 | 1234 | 5.000 | 5.085 | 0.955 | 0.000193 | 84.054 |
Parkinson’s disease dataset | ||||||||||||
 RFE | 76.441 | 75.133 | 1.308 | 0.384 | 0.480 | 0.426 | 376.000 | 144.783 | 0.177 | 0.689 | 0.00145 | 75.133 |
 RFS | 76.484 | 75.000 | 1.484 | 0.629 | 0.557 | 0.537 | 224.000 | 1.474 | 0.158 | 0.706 | 0.00108 | 75.000 |
BreastEW dataset | ||||||||||||
 RFE | 95.000 | 94.000 | 1.000 | 0.948 | 0.937 | 0.941 | 15.000 | 0.142 | 0.099 | 0.990 | 0.00050 | 94.000 |
 RFS | 95.000 | 93.000 | 2.000 | 0.938 | 0.928 | 0.932 | 27.000 | 0.090 | 0.008 | 0.989 | 0.00583 | 93.000 |
Dermatology erythemato-squamous diseases dataset | ||||||||||||
 RFE | 97.541 | 96.997 | 0.544 | 0.972 | 0.960 | 0.963 | 17.000 | 0.062 | 0.001 | 0.997 | 0.001073 | 96.997 |
 RFS | 89.860 | 87.725 | 2.135 | 0.837 | 0.821 | 0.815 | 11.000 | 0.094 | 0.002 | 0.985 | 0.002819 | 87.725 |