Research Article
Predicting Breast Cancer Based on Optimized Deep Learning Approach
Table 3
The performance of applying regular ML models and DL model with selected features by correlation matrix.
| Approaches | Model | CV performance | Testing performance | AC | PR | RE | FM | AC | PR | RE | FM |
| Regular ML approach | DT | 94.4 | 94.98 | 94.51 | 94.56 | 92.11 | 92.11 | 92.11 | 92.1 | KNN | 89.85 | 90.63 | 89.85 | 89.51 | 86.67 | 87.4 | 86.67 | 86.17 | NB | 81.84 | 82.38 | 81.84 | 81.01 | 83.68 | 84.33 | 83.68 | 83.0 | RF | 97.01 | 96.74 | 96.75 | 96.68 | 94.04 | 94.05 | 94.04 | 94.03 | SVM | 94.73 | 94.94 | 94.73 | 94.66 | 93.86 | 93.85 | 93.86 | 93.84 |
| DL approach | The optimized deep RNN | 97.92 | 97.77 | 97.79 | 97.78 | 95.18 | 95.44 | 95.18 | 95.21 |
|
|