Efficient Machine Learning Techniques for Improving the Prediction of Chronic Disease

Abstract

Chronic diseases are often considered as a major source of concern and a threat to public health on a global scale. Chronic diseases such as Chronic Kidney Disease (CKD), Chronic Diabetes Mellitus (CDM) and Cardiovascular Disease (CVD) are severe chronic diseases that claim millions of lives each year. Each of these individual disorder is considered as potential risk factor for the other two. As a result, great effort is being done to reduce the chance of developing these diseases as a preventive measure. The supervised as well as unsupervised machine learning algorithms applied in early disease prediction are found computationally intensive which frequently overfit and underperform in terms of accuracy because they should analyse the vast amount of clinical information till the convergence of the model. The primary objective of the research work is to propose feature selection algorithms to reduce the dimensionality of the chronic disease dataset by eliminating the redundant and irrelevant features to increase the accuracy of the prediction. newline