Date of Graduation

5-2026

Document Type

Thesis

Degree Name

Bachelor of Science in Mathematics

Degree Level

Undergraduate

Department

Mathematical Sciences

Advisor/Mentor

Robinson, Samantha

Committee Member

Maddox, Kyle

Second Committee Member

Dingman, Shannon

Third Committee Member

Mitchell, Marc

Abstract

Recently noted, Huang et al. (2023), machine learning (ML) models, while offering great advantages over traditional statistical predictive modeling methods, are less explored in the analysis of survival and other similar time-to-event predictive data modeling. ML methods such as neural networks offer a great deal of promise but need to be further explored to investigate their comparative power in predicting survival outcomes. Focusing specifically on survival analysis in adult and pediatric bone cancer patients, traditional methods, like shown in Emmert-Streib and Dehmer (2019), will be shown with machine learning models using methods in Hothorn, Hornik, and Zeileis (2006). In this project, research questions include: What relationship does age have with survival of patients at different stages of bone cancer?

Once models have been created, using both traditional statistical methodology and ML, the accuracy of performance and computational speed will be compared using data from the National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) Program database as well as available regional cancer registry data.

Keywords

Survival analysis; statistics; pediatric; conditional inference trees

Download

Share

COinS