Ensemble Learning for Diabetes Detection: A Voting-Based Classifier Approach

Disasters often lead to diseases affecting both adults and children, causing physical and mental illness. Diabetes, a rapidly rising chronic illness, is difficult to diagnose early without regular doctor visits. The International Diabetes Federation reports that diabetes affects 1 out of 11 individuals worldwide, with a higher prevalence in middle to low-income nations. The increasing rate of diagnosis underscores the importance of early detection. In this study, classifiers like SVM, XGB, LR, RF, and LDA were tested for accurate prediction, with a voting classifier achieving 90.10% accuracy using the PIMA Indian dataset.