Deep learning approach for Diabetes Risk Analysis

Abstract

Diabetes is a chronic disease that is affecting more and more people around the world because of high glucose levels. The prevalence and incidence rates have been very much to increase year-on-year. If left untreated, the health problems associated with diabetes in most of the organs of the body can be devastating. The main purpose of which is to predict in patients with diabetes mellitus. Various classification algorithms such as Support Vector Machine(SVM), K-nearest Neighbor (KNN), Decision tree, Naïve Bayes(NB), can be used on the Pima Indians Diabetes Database (PIDD), downloaded from the UCI repository of machine learning and compared with an approach of deep learning used to predict the development of diabetes mellitus. The accuracy is measured based on machine learning(ML) classification algorithms with a comparison of deep learning(DL) model.