Research Article
A Hybrid Intelligent System Framework for the Prediction of Heart Disease Using Machine Learning Algorithms
Table 10
Excellent performance metrics results and best classifiers with feature selection algorithms for n = 6 with 10-fold CV.
| ā | Best performances evaluation metrics and best classifiers |
| FS | The best accuracy (%) and classifier | The best specificity (%) and best classifier | The best sensitivity (%) and classifier | The best MCC and classifiers | The best AUC and classifiers | The best processing time(s) and classifiers | Relief | 89 logistic regression with C = 100 | 98 logistic regression with C = 100 | 100 ANN (MLP) with 16 | 89 logistic regression with C = 100 | 88 logistic regression | 14.134 SVM ( RBF) with C = 100, G = 0.0001 | mRMR | 84 Naive Bayes | 100 SVM (linear) with C = 100, = 0.0001 | 98 ANN (MLP) with 20 | 83 Naive Bayes | 84 Naive Bayes | 1.121 random forest | LASSO | 88 SVM( RBF) with C = 100, = 0.001 | 97 logistic regression with C = 100 | 78 Naive Bayes | 88 SVM (RBF) with C = 100, = 0.001 | 89 SVM (RBF) with C = 100, = 0.001 | 0.005 SVM( linear) C = 100, = 0.001 |
|
|