Research Article
A New Feature Ensemble with a Multistage Classification Scheme for Breast Cancer Diagnosis
Table 8
The evaluation metrics of each classifier evaluated by elevenfold cross-validation for the two-stage study.
| Classifier | SNS | SPC | PPV | NPV | FPR | FNR | FDR | FOR | ACC |
| LLC | 86.49 | 92.11 | 84.92 | 91.78 | 7.89 | 13.51 | 15.08 | 8.22 | 86.80 | FLDA | 87.45 | 92.18 | 85.25 | 91.76 | 7.82 | 12.55 | 14.75 | 8.24 | 87.51 | LDC | 84.68 | 90.29 | 81.66 | 89.41 | 9.71 | 15.32 | 18.34 | 10.59 | 85.56 | SVM (‘p’, 1) | 84.46 | 90.13 | 80.30 | 89.51 | 9.87 | 15.54 | 19.70 | 10.49 | 84.12 | Decision tree | 75.78 | 87.64 | 74.93 | 88.00 | 12.36 | 24.22 | 25.07 | 12.00 | 75.79 | Random forest | 77.51 | 88.29 | 77.38 | 88.38 | 11.71 | 22.49 | 22.62 | 11.62 | 77.48 | Naïve Bayes | 67.81 | 83.86 | 71.38 | 84.75 | 16.14 | 32.19 | 28.62 | 15.25 | 68.33 | kNN () | 42.13 | 70.51 | 40.73 | 70.81 | 29.49 | 57.87 | 59.27 | 29.19 | 42.12 |
|
|