Research Article
A New Feature Ensemble with a Multistage Classification Scheme for Breast Cancer Diagnosis
Table 5
The evaluation metrics of each classifier evaluated by elevenfold cross-validation for the one-stage study.
| Classifier | SNS | SPC | PPV | NPV | FPR | FNR | FDR | FOR | ACC |
| LLC | 85.47 | 92.74 | 85.47 | 92.74 | 7.26 | 14.53 | 14.53 | 7.26 | 85.47 | FLDA | 81.72 | 90.86 | 81.72 | 90.86 | 9.14 | 18.28 | 18.28 | 9.14 | 81.72 | LDC | 80.80 | 90.40 | 80.80 | 90.40 | 9.60 | 19.20 | 19.20 | 9.60 | 80.80 | SVM (‘p’, 1) | 77.14 | 88.57 | 77.14 | 88.57 | 11.43 | 22.86 | 22.86 | 11.43 | 77.14 | Decision tree | 77.50 | 88.75 | 77.50 | 88.75 | 11.25 | 22.50 | 22.50 | 11.25 | 77.50 | Random forest | 75.86 | 87.93 | 75.86 | 87.93 | 12.07 | 24.14 | 24.14 | 12.07 | 75.86 | Naïve Bayes | 65.42 | 82.71 | 65.42 | 82.71 | 17.29 | 34.58 | 34.58 | 17.29 | 65.42 | kNN () | 42.13 | 71.06 | 42.13 | 71.06 | 28.94 | 57.87 | 57.87 | 28.94 | 42.13 |
|
|