Integrating Machine Learning for Accurate Prediction of Early Diabetes

Abstract

In the current world, where diabetes is day by day becoming a very common and fatal disease, it's important that proper measures be taken in order to deal with it. As per the studies, early prediction of diabetes can lead to improved treatment to avoid further complications of the disease, and in order to do so efficiently, machine learning techniques are a great deal. In this study, various factors are taken into consideration, like blood pressure, pregnancy, glucose level, age, insulin, skin thickness, and diabetes pedigree function, which together can be useful to predict whether a person has a risk of developing diabetes or not and help society with the early diagnosis of diabetes. This model is trained using three main classification algorithms, namely support vector, random forest, and decision tree classifiers. The prediction results of each of the classifiers are summarized in this study, and the decision tree gives 78.89% accuracy.