Early diabetes prediction via C4.5 decision tree model: a machine learning approach

Chronic diseases pose significant health challenges, and early detection is crucial for effective management and prevention of complications. This study leverages the C4.5 decision tree algorithm to develop a predictive model for early disease detection. Data from 520 patients at a medical institution, encompassing various demographic and clinical indicators, were utilized. The C4.5 algorithm constructs a decision tree to classify patients based on their disease risk. Model evaluation using metrics such as accuracy, precision, and recall demonstrated superior performance compared to other regression models, including ridge regression, best subset regression, and logistic regression. The C4.5 decision tree model achieved an overall accuracy of 94.23%, effectively balancing precision and recall for both positive and negative cases. This study highlights the efficacy of the C4.5 decision tree in capturing complex patterns in medical data, providing a robust tool for early disease prediction and potentially aiding in optimized patient care and resource allocation.