DETECTING DISEASES OF VARIANT NATURE IN HUMANS BY ENHANCED ALGORITHM USING SVM

Abstract: The healthcare organisation creates a massive amount of patient data, which may be analysed in a variety of ways. As a result, with the assistance of a machine learning, we developed a prediction system that can identify many diseases at the same time. We have focused on various diseases: heart disease, liver disease, diabetes, etc however many more diseases may be included in the future. The user must enter numerous illness parameters, and the system will determine whether the person has the diseases or not. Support vector machines with adaptivity were utilised to identify numerous illnesses. The goal was to offer an adaptive SVM-based diagnostic technique that was automated, rapid, and versatile. To improve outcomes, the bias value in traditional SVM was changed. The suggested classifier produced 'if-then' rules. Using the recommended technique, several diseases were detected, as well as increased categorization rates. The key emphasis of future research should be the development of more effective ways for changing the bias value in classical SVM.

Keywords: Diseases Prediction System, Supervised Machine Learning, Classification, Prediction, Support Vector Machine, Health Care Analysis.