MACHINE LEARNING-BASED DIABETES RISK ASSESMENT IN PRIMARY HEALTHCARE

Abstract

Diabetes is a common health problem that many people have. It is a global health issues that usually prolonged in a 
patient for an entire life. It increases the risk of long-term complications including heart disease, and kidney failure 
among others. If we cannot take proper steps to diagnose diabetes at an early stage, eventually we have to face serious 
health issues. People might live longer and lead healthier lives if this disease is detected early. Therefore, machine 
learning classification algorithms such as Decision Trees, Naive Bayes, Support vector machine, and Random Forest 
to detect diabetes have been used in this study. This information includes things like age, polyuria, obesity, sudden 
weight loss, and weakness. After using all the patient records, we are able to build a machine learning model to 
accurately predict whether or not the patients in the dataset have diabetes. We found that these Machine Learning 
Algorithms are good at predicting who might get diabetes early. Machine learning algorithm showing the highest 
accuracy is selected for further development. The Machine Learning tells us who might get diabetes in the future. This 
helps doctors give advice and support to people so they can stay healthy and not get sick with diabetes. The results of 
this study suggest that applying ML-based classification may predict diabetes accurately and detect the diabetes
very efficiently.