A Survey On Diabetes Prediction Using Machine Learning Techniques

Abstract

In the recent years, most of the people in the world are affected with blindness, kidney failure, amputations, heart failure and stroke. It is found that all of these are caused by most probably by diabetes disease. While food eaten is being converted into sugars or glucose, the pancreas releases insulin, which servers as an important key to open the cells of the human body and it allows the glucose to enter and it allows to use the glucose for energy. If the insulin is not properly created or released, several things may go wrong, which in turn onset of diabetes. This disease is classified as Type 1, Type 2 which are commonly occurred and Gestational diabetes occurred during pregnancy. To detect and predict this disease in the earlier stage will help the people to get relief from it. Different machine learning approaches have been proposed to predict the diabetes using data sets like PIMA dataset. This paper includes the comparative study of performance metrics for various machine learning approaches have been used.