Medical Diagnoses: Heart Disease Prediction

The successful and continuously increasing applications of machine learning (ML) and data mining (DM) in the wide range of fields such as e-businesses, public or private sectors, marketing and retails has led the ML and DM applications in other industries and health sectors. In the health sector the data is information wise too rich but not knowledge rich. It can be due to the lack of effective tools and research study in this era. In this research work, an efficient model is proposed based on the machine learning and data mining tools to predict a particular medical disease that is the heart disease. The solution for the problem which is addressed that data is not knowledge-rich, is presented in this paper. Waikato Environment and Python is used for to analyze the data, so that an automation system can be formed. Several experiments are to be implemented to compare the performance through Supervised Learning and Unsupervised Learning. Bayesian learning (Naïve Bayes), Tree’s algorithm (Decision Tree), and Neural Networks (MLPNN) are applied on the same dataset. Comparatively, Decision Tree gives best accuracy when compared with Naïve Bayes, ANN performs more effective when compared with others two Bayesian algorithms (Naïve Bayes) and Tree’s algorithm (Decision tree).