Capstone Projects

Early Detection of Diabetes Using Machine Learning Models

Program: Data Science Master's Degree
Host Company: Case Study with support from United Health Care Group
Location: Minnetonka, Minnesota (onsite)
Student: Taylor Czaplewski

The objective of this project was to predict if an individual has diabetes. The history of diabetes was examined to gain a better understanding of its origins and what is currently known about the disease. It also looks into current studies that are being done to help find better ways to manage symptoms and even cure the disease altogether. This project then utilizes data from the Nation Health Nutrition Examination Survey to create a backward stepwise regression model to predict whether individuals are likely to have diabetes or not. The results of this model can then be used to help detect diabetes in an earlier state to allow individuals to live a healthier lifestyle to prevent a diabetes diagnosis or delay the diagnosis.