Comparative Analysis for the Prediction of major Diseases using Supervised and Hybrid Machine Learning Algorithms

Abstract: In the few recent years, common major diseases have emerged together of the foremost common causes of deaths worldwide. As the changing in lifestyle, food habits, working cultures etc, has significantly contributed to the present issues related to health across world-wide including the developed, underdeveloped and developing countries, a challenge to the medical science to overcome from this situation[1]. As per as WHO (World Health Organization) Global Health Estimates report is concerned, an estimated 74% of all deaths were noncommunicable diseases globally, 3 out of 10 major diseases are communicable. In this paper, we have taken 5 major diseases (Ischaemic Heart Disease (IHD) with Stroke, Chronic Kidney Disease (CKD), Diabetes Mellitus (DM) including BP, Chronic Liver, and Cancer) among the top 10 deadliest diseases[2]. All these major diseases can be curable with proper diagnosis and early detection. The purpose of this paper is to establish some Machine Learning supervised algorithms with some hybrid approach for better comparative analysis and predict for the particular disease at an early stage with a greater accuracy level. The outcome of this paper also justify that the hybrid algorithm model has better processing, performance with more accuracy outputs so as to help the medical and healthcare sector in the early stage disease prediction.

Keywords: Algorithm, Classifier, Machine Learning, Predictive Analysis.