Support Vector Machines-based Heart Disease Diagnosis using Feature Subset, Wrapping Selection and Extraction Methods

Heart disease is one of the leading causes of human death and in the absence of an accurate diagnosis, there are limitations to beat it. In this research, an automatic diagnostic methodology for clinical heart disease is presented. The proposed approach computes the most relevant feature subset by taking advantage of feature selection and extraction techniques. To accomplish the feature selection, two algorithms (Mean Fisher based feature selection algorithm(MFFSA) and accuracy based feature selection algorithm(AFSA)) are presented. The selected feature subset is then further refined through the feature extraction technique i.e., principle component analysis. The proposed technique is validated over Cleveland, Hungarian, Switzerland and combination of all of them. In order to classify a human as a heart disease patient(HDP) or a normal control subject(NCS), Radial basis function kernel-based support vector machines are used. The proposed methodology is evaluated through accuracy, specificity and sensitivity metrics.