Diabetes Prediction Using Machine Learning

Ritik

Abstract: Diabetes is one of most common diseases worldwide that occurs when your blood glucose is too high. Blood glucose is your main source of energy and comes from the food you eat. Annually it costs a lot of money to take care people with diabetes. Thus the most important issue arises to predict diabetes to be very accurately. One of these methods is using machine learning model. So in this paper, we used a machine learning to predict whether a person is diabetic or not. To build the onset Diabetes predictor by machine learning we used PIMA (Indian Diabetes Datasets). After analyzing dataset and training the machine learning model using Random Forest algorithm, the average probability was 0.75 and the average accuracy was coming to predict if a person is diabetic or not was 77.3%. but to get more accurate result, we used XgBoost (machine learning library) which in result able to give more accurate details with high accuracy of 85.24%.

Keywords: Diabetes, Machine Learning, Algorithm, prediction