Research Article
Hybrid Feature Selection Based Weighted Least Squares Twin Support Vector Machine Approach for Diagnosing Breast Cancer, Hepatitis, and Diabetes
Table 4
Performance comparison of proposed WLSTSVM model.
| Performance | Pima Diabetes | Hepatitis | Breast Cancer | Evaluation | Original features | Reduced features | Original features | Reduced features | Original features | Reduced features | Parameters | 8 | 5 | 19 | 12 | 9 | 5 |
| Accuracy | 75.67% | 89.71% | 86.67% | 87.50% | 95.47% | 98.55% | Sensitivity | 87.09% | 96.94% | 91.67% | 95.14% | 98.00% | 100% | Specificity | 69.57% | 83.11% | 75.54% | 82.96% | 93.55% | 97.37% | Geometric mean | 77.84% | 89.76% | 83.22% | 88.84% | 95.75% | 98.68% |
|
|