Prediction system for heart disease using Naive Bayes and particle swarm optimization

Heart attack disease is major cause of death anywhere in world. Data mining play an important role in health care industry to enable health systems to properly use the data and analytics to identify impotence that improves care with reduce costs. One of data mining technique as classification is a supervised learning used to accurately predict the target class for each case in the data. Heart disease classification involves identifying healthy and sick individuals. Linear classifier as a Naive Bayes (NB) is relatively stable with respect to small variation or changes in training data. Particle Swarm Optimization (PSO) is an efficient evolutionary computation technique which selects the most optimum features which contribute more to the result which reduces the computation time and increases the accuracy. Experimental result shows that the proposed model with PSO as feature selection increases the predictive accuracy of the Naive Bayes to classify heart disease.