Date of Award

5-2025

Document Type

Honors Thesis

Degree Name

Bachelor of Science

Department

Information Science

Advisor/Committee Chair

Unal Tatar

Abstract

Generative AI is becoming increasingly integrated within various parts of our society, with education being an important area. Students can utilize generative AI to assist with their homework and have a personalized and curated tutor. However, current models make it difficult for students to fully engage in the learning process as these models immediately give students their answers without showing the critical thinking or problem-solving skills involved in the problem. As AI slowly becomes more entangled in education, it is crucial to consider how Generative AI models can assist students in approaching their studies with a problem-solving mindset. This research narrows down on STEM students, particularly in technology-aligned classes as the problem-solving techniques vary between different disciplines. This study aims to understand how Generative AI models can be customized to help students achieve a problem-solving mindset when approaching difficulties in their studies. This study will create a customized GPT model, utilizing features that would enhance an LLM’s capabilities for better student success. An assessment criteria framework will be created to compare a standard GPT model versus this study’s customized GPT to view the effects on student success.

Creative Commons License

Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License
This work is licensed under a Creative Commons Attribution-NonCommercial-Share Alike 4.0 International License.

Available for download on Saturday, May 15, 2027

Share

COinS