Probabilistic decision support system using machine learning techniques : A case study of Cardiovascular diseases

Abstract

Diseases are an irregular event that affects single or multiple body parts of a human being. Different kinds of diseases are growing day by day in terms of lifestyle and heritage. Among all the ones, it turns out that heart disease is the most prevalent disease, and the consequence of this disease is dangerous relative to all other situations. In this paper, initially the author calculates immanent discreteness between two probability separations, based on information theory or relative entropy using Kullback-Leibler divergence (KL divergence). It can be used as shortcuts in the calculation of other widely used methods, such as mutual information for feature selection before modeling, and cross-entropy used as a loss function for many different classifier models. Seven computational intelligence techniques named logistic regression (LR), support vector machine (SVM), naïve bayes (NB), random forest (RF), and k-nearest neighbor (K-NN) have been applied and a comparative study had been drawn. The heart disease dataset is taken from the Department of Cardiology, Excelcare Hospitals, Guwahati, Assam and the analysis of performance in each technique has been evaluated.