Document Type
Thesis
Degree
Master of Science
Major
Computer Science
Date of Defense
7-18-2025
Graduate Advisor
Dr. Badri Adhikari
Committee
Dr. Badri Adhikari
Dr Azim Ahmadzadeh
Dr. Sambriddhi Mainali
Abstract
Argumentative writing is a critical skill that strengthens students’ reasoning, communication, and analytical abilities. However, maintaining a clear and organized argument structure while writing can be challenging. Argument maps — visual diagrams which explicitly show an argument’s structure — have been shown to improve students’ writing, but are rarely used outside of the planning stage of an essay due to the time and effort required to create them. Automatically generating argument maps from student essays helps students to evaluate the structure of their argument as they write and makes identifying unsupported claims visible. To evaluate whether large language models (LLMs) can generate accurate argument maps from student essays, this study uses examples from The Learning Lab’s essay scoring competi- tion dataset. The essays are grouped by topic and by holistic score, which measures the writer’s mastery of source-based writing. An LLM was instructed using a few-shot prompting pattern and provided essays and argument maps as examples. The LLM’s capability was evaluated qualita- tively by comparing the generated argument map with the written essay to determine how well the argument map captured the overall argument, evidence used, and relationship between the argument’s components. The application presented in this paper is a framework for generating argument maps that accurately represent student-written essays. Argument maps reflect the content of a given essay and logically structure the evidence within. Edge cases such as prompt hijacking, single-phrase inputs, and intentionally contradictory arguments cause the model to hallucinate results. Due to these issues, further development would be needed before academic use. The model’s prompt can be edited through the interface to allow for ease of testing.
Recommended Citation
Chalmers, Alexis R., "A Web Application for Generating Argument Maps for Essays Using LLMs" (2025). Theses. 496.
https://irl.umsl.edu/thesis/496