Uncovering the Impact of Untreated Diabetes via Machine Learning Approach

Determining the impact of untreated diabetes poses challenges with an estimated one in four cases remaining undiagnosed. Diabetes mellitus, a group of metabolic disorders, arises when blood glucose levels or blood sugar are elevated due to insufficient insulin production, cellular response or both. We have diagnosed untreated diabetes to predict the level of diabetes in this paper. In this paper we tested various machine learning algorithms to improve the precision of disease prediction while automating health care systems. The classifier and boosting techniques include Ada Boost, Logistic Regression, Decision Tree, Extreme Gradient Boosting, Gaussian Naive Bayes, Support Vector Machine, Random Forest, and Cat Boost. Decision Tree emerged as the top-performing algorithm, achieving an accuracy of 0.8306 in classifying diabetes health markers, followed by Logistic Regression having 82.97. It's common to evaluate the performance of machine learning models using multiple metrics to get a comprehensive understanding of their effectiveness.