Shi and Zhou Team to Produce Book Chapter on Diabetes Prediction

Turner College economist Wen Shi and TSYS School computer scientist Yi Zhou recently teamed with Yuan Tian and Ying Zhou of Auburn University and Chuan Wang of Vanderbilt University on a book chapter titled "Enhancing Diabetes Prediction with Advanced Machine Learning Techniques." Their chapter leverages data from the Framingham study to investigation the utilization of machine learning models in the realm of diabetes prediction. The researchers analyze multiple machine learning algorithms, including XGBoost, AdaBoost, Random Forest and LightGB, all of which were selected for their proven capabilities in handling complex datasets and providing accurate predictions in various domains. The results indicate that AdaBoost and LightGB are the most consistent and reliable algorithms, particularly in cases where class distribution is skewed. As they conclude, the effective deployment of these machine learning models in diabetes prediction showcases the harmonious integration of technology with healthcare. The chapter appears in Scientific Computing and Bioinformatics and Computational Biology (Springer, 2005), which is edited by Douglas Hodson, Michael Grimaila, and Torrey Wagner of the U.S. Air Force Institute of Technology, Hamid Arabnia of the University of Georgia, and Leonidas Deligiannidis of Wentworth Institute of Technology.