Diabetes Prediction Using Machine Learning and Flask

N Kushal Kumar Raju^1*, Keshav Krishnamurthy¹, Bhuvanagiri Prahal Bhagavath², Nathan Shankar², A. M. Janani³, N Avinash², Aditya Ray¹ and P. Mahalakshmi⁴

¹Electronics and Instrumentation Engineering, School of Electrical Engineering, Vellore Institute of Technology, Vellore, Tamil Nadu India.

²Electrical and Electronics Engineering, School of Electrical Engineering, Vellore Institute of Technology, Vellore, Tamil Nadu India.

³Computer Science and Engineering, Sri Sairam Engineering College, Chennai, Tamil Nadu India.

⁴School of Electrical Engineering, Vellore Institute of Technology, Vellore, Tamil Nadu India.

Corresponding Author E-mail: pmahalakshmi@vit.ac.in

Abstract: Diabetes is one of the costliest chronic diseases, it is a metabolic disorder in which a patient has excessive blood sugar levels due to the body's inability to create enough insulin, and it can also cause long-term harm to the heart, blood vessels, eyes, kidneys, and nerves. Adults with diabetes are twice as likely as non-diabetics to have a heart attack or stroke. Despite its massive impact on the global population, no kind of diabetes has a cure. Although most medications help patients manage their symptoms to some extent, diabetics nevertheless suffer several long-term health concerns. So, if we are able to predict diabetes early, we could control it and it can be done by using Machine learning techniques. Our work aim is to predict if the patient has diabetes using Machine learning techniques and the ensemble method. We will be using four algorithms which are SVM, KNN, Logistic Regression, and Random Forest classifier and we would also compare all four models to check which model is giving the best accuracy and link our best model to a web app that could predict if the patient has any chances of having diabetes.

Keywords: Ensemble; KNN; Logistic Regression; Machine learning; Random Forest; SVM

[ HTML Full Text]

Back to TOC