ABSTRACT_ Data is a valuable resource in this digital age, and there was a tonne of data being generated across all fields. Health care industry data comprises contains information about the patient as well as information on the ailment. The use of machine learning and medical data will assist us in doing data analysis to uncover hidden illness patterns and develop individualized treatments for the patient and utilized to forecast the illness. A generic architecture for illness prediction has been presented in this work in the field of medicine. Using an improved SVM-Radial bias kernel method, this system was tested using a smaller set of features from the Chronic Kidney Disease, Diabetes, and Heart Disease dataset. It was also compared to other machine learning methods such as SVM-Linear, SVMPolynomial, Random forest, and Decision tree in R studio. All of these machine learning algorithms have been assessed based on their performance