A Study on Non-invasive Diabetes Causing Variables and Their Covariance Relationship in Diabetes Prediction Using Machine Learning Algorithms

Diabetes is a chronic disease that affects millions of people worldwide, and its early detection and management are crucial for preventing serious health complications. In recent years, machine learning has emerged as a promising approach for diabetic prediction and management. In this project, we develop a system for accurately predicting diabetes using machine learning models such as random forest classifier, support vector machine, XGBoost classifier, and decision tree algorithm. We use a dataset containing various features related to patients, including age, body mass index, blood pressure, and glucose levels, to train and evaluate our models. Our results show the most important features that contribute to the prediction accuracy, providing insights for personalized diabetes management. Experimental results determine the adequacy of the designed system with an achieved accuracy of 79% using random forest.