A Comparative Analysis of Machine Learning Models for Early Prediction of Diabetes

Abstract: Diabetes mellitus is a chronic metabolic disorder and a major global health concern. Early prediction is essential to reduce complications and healthcare burden. This paper presents a comparative analysis of Linear Regression, Exponential Regression, and Logistic Regression models for early diabetes prediction using clinical data. A dataset of 761 patient records was used, and model performance was evaluated using accuracy, precision, recall, F1-score, and confusion matrix under 10-fold cross-validation. Logistic Regression achieved the highest validation accuracy of 92.68% and an F1-score of 89.77, outperforming the other models. The results indicate that Logistic Regression is the most reliable model for early diabetes detection in clinical settings.

Keywords: Diabetes Prediction, Machine Learning, Linear Regression, Logistic Regression, Exponential Regression