Computational intelligence technique for early diagnosis of heart disease

Computational intelligence plays a vital role in heart disease diagnosis. Researchers have been using several intelligence techniques to improve the heart disease diagnosis accuracy. Data mining plays an important role in the field of heart disease prediction. Classification is a supervised learning in data mining, used to accurately predict the target class for each case in the data. Naive bayes classifier belongs to a family of linear classifiers and classifier learning is relatively stable with respect to small changes in training data. Heart disease is a leading cause of death over the decade. Heart disease classification involves to identify healthy and sick individuals. In this paper, we attempted to increase the predictive accuracy of the naïve bayes to classify heart disease data. We used a discretization method and genetic search to remove redundant features. Genetic search is used for optimization problem. Finally we performed a comparison with other approaches that have tried to improve the accuracy of naïve bayes classifier for heart disease classification.