Project Overview
For my project, I plan to develop a fully operational website equipped with a large language model (LLM) as a way to help students seamlessly brainstorm and test different educational paths. The aim of this project is to vastly improve students’ preparation for registration and therefore their overall academic experience at Georgetown. In the system I hope to create, students could prompt the model with a few different majors and minors and receive a variety of 4-year academic plans as output. Beyond this, students could interact with the LLM by discussing their interests to learn more about the different choices at their disposal. Ultimately, such a system would be like having a virtual dean at your fingertips.
From my personal experience and discussions with my peers, I have discovered that many students miss out on taking courses that they are interested in due to their difficulty interacting with myDegree and navigating Georgetown’s course descriptions and prerequisites. By training the model on different academic departments’ missions, their majors, minors, and certificates, as well as the subsequent course requirements, I can equip the LLM to talk students through different plans tailored to their specific interests. Additionally, incorporating data from the career center could help students consider potential career paths and how they relate to the classes that they plan on taking.
To be clear, this system would not be created to replace deans. Rather, the LLM would complement the work of Georgetown deans, allowing students to experiment and research course options without the time constraint of an in-person meeting. Ultimately, using natural language as a versatile interface for registration preparation will vastly improve upon the rigid myDegree system that struggles with ease-of-use and freedom of interaction.
Team
Andrew Bank
College of Arts and Sciences – Computer Science
