Research Article
Hybrid Feature Selection Based Weighted Least Squares Twin Support Vector Machine Approach for Diagnosing Breast Cancer, Hepatitis, and Diabetes
Table 7
Comparison of predictive accuracies of proposed and other classifiers for Breast Cancer.
| | Reference | Approach | Predictive accuracy |
| | This study | HFS + WLSTSVM (10 × CV) | 98.55% | |
Karabatak and Ince [34] | AR + NN | 97.4% | |
Chen et al. [31] | GA | 96.99% | |
Sousa et al. [51] | Discrete particle swarm optimization | 94% | | Akay [29] | FS + SVM (train: 75%-test-25%) | 99.51% | | Quinlan [24] | C4.5 (10 × CV) | 94.74% | |
Hamilton et al. [26] | RIAC (10 × CV) | 95.00% | | Ster and Dobnikar [27] | LDA (10 × CV) | 96.80% | |
Polat and Güneş [33, 40] | LS-SVM | 98.53% | |
Abonyi and Szeifert [52] | Supervised fuzzy custering | 95.57% | |
Goodman et al. [53] | AIRS | 97.20% | | Bennett and Blue [54] | SVM (5 × CV) | 97.20% | | Şahan et al. [55] | Fuzzy AIS - KNN (10 × CV) | 99.14% |
|
|
