Date of Award
12-2024
Degree Name
Master of Arts
Department
Family and Consumer Sciences
First Advisor
Ping Ouyang, Ph.D.
Second Advisor
Ya Zhang, Ph.D.
Third Advisor
Arezoo Rojhani, Ph.D.
Access Setting
Masters Thesis-Open Access
Abstract
This research analyzes Behavioral Risk Factor Surveillance System data from 2022 with three machine learning models, Neural Network, Random Forest, and Support Vector Machine, to determine and compare the models’ accuracy rate of diabetes prediction. Principal Component Analysis was employed to determine principal factors of diabetes risk. Results show that the Neural Network is most accurate at 90.09 percent at predicting diabetes. Factors increasing diabetes risk include alcohol consumption, smoking, and self-perceived mental and physical health.
Recommended Citation
Weaver, Addie, "Machine Learning and Evaluation of Diabetes Risk Using 2022 BRFSS Data" (2024). Masters Theses. 5447.
https://scholarworks.wmich.edu/masters_theses/5447
Included in
Analytical, Diagnostic and Therapeutic Techniques and Equipment Commons, Endocrinology, Diabetes, and Metabolism Commons
