Skip to main content

Table 2 SVM diagnosis for benchmark data under 5-fold cross validation

From: Diagnostic biases in translational bioinformatics

Algorithm Accuracy ± std (%) Sensitivity ± std (%) Specificity ± std (%) NPR ± std (%) PPR ± std (%)
   BreastIBC data    
SVM-linear 74.56 ± 04.52 97.14 ± 06.39 16.67 ± 23.67 NaN 75.70 ± 06.52
SVM-rbf 72.56 ± 03.63 100.0 ± 00.00 00.00 ± 00.00 NaN 72.56 ± 03.63
SVM-quad 74.56 ± 04.52 97.14 ± 06.39 16.67 ± 23.67 NaN 75.70 ± 06.52
SVM-rbf2 72.83 ± 10.92 85.71 ± 14.29 40.00 ± 09.13 63.33 ± 34.16 78.65 ± 05.88
SVM-mlp 45.67 ± 18.09 48.10 ± 22.99 40.00 ± 09.13 25.67 ± 14.02 65.33 ± 12.16
   HCC data    
SVM-linear 94.02 ± 01.43 95.81 ± 03.83 92.42 ± 05.21 96.17 ± 03.50 92.39 ± 05.00
SVM-rbf 52.00 ± 00.75 100.0 ± 00.00 00.00 ± 00.00 NaN 52.00 ± 00.75
SVM-quad 82.05 ± 10.66 77.00 ± 10.77 87.52 ± 12.21 77.87 ± 10.32 87.38 ± 11.89
SVM-rbf2 89.90 ± 04.32 92.86 ± 08.75 87.17 ± 06.48 93.60 ± 07.25 87.33 ± 05.32
SVM-mlp 51.87 ± 10.96 46.00 ± 15.80 58.29 ± 10.06 50.43 ± 08.72 53.44 ± 14.68
   Kidney data    
SVM-linear 90.23 ± 02.35 96.84 ± 03.07 44.07 ± 06.63 71.46 ± 16.90 92.38 ± 00.71
SVM-rbf 87.48 ± 00.44 100.0 ± 00.00 00.00 ± 00.00 NaN 87.48 ± 00.44
SVM-quad 87.47 ± 01.70 94.47 ± 01.20 17.47 ± 07.89 50.00 ± 21.21 89.21 ± 00.80
SVM-rbf2 87.48 ± 00.44 100.0 ± 00.00 00.00 ± 00.00 NaN 87.48 ± 00.44
SVM-mlp 53.39 ± 06.79 54.32 ± 07.79 46.92 ± 10.08 13.02 ± 02.95 87.67 ± 02.47