Using chaotic adaptive PSO-SVM for heart disease diagnosis

In this paper, the application of support vector machine (SVM) approach based on the statistics-learning theory of structural risk minimization in heart disease diagnosis. Aiming at the blindness of man made choice of parameter and kernel function of SVM, a chaotic adaptive particle swarm optimization (CAPSO) method is applied to select parameters of SVM in the paper and genetic characteristics of the subset of choices (GA_FSS) is applied to reduce large dimensions and improve greatly the accuracy of classification. The experimental results on heart disease diagnosis problem show that CAPSO-SVM .classifier algorithm is effective and correct.