Project Director

Liang, Yu

Department Examiner

Wigal, Cecelia; Howell, Roland

Department

Dept. of Computer Science and Engineering

Publisher

University of Tennessee at Chattanooga

Place of Publication

Chattanooga (Tenn.)

Abstract

DataPeer is a web application that combines open-ended high agency (OHA) and structured low-agency (SLA) interaction paradigms to support human-LLM collaboration in exploratory data analysis. DataPeer integrates a large language model (LLM) with a React frontend and FastAPI backend, allowing both qualitative and quantitative analysis of user-uploaded CSV datasets through a chat-based interface. Users can attach datasets and provide natural language queries, while the LLM provides data analysis to the user. This thesis investigates the integration of LLMs into data analysis workflows and addresses gaps in interaction design and user agency in LLM-driven data tools by offering a responsive interface useful for both experts and non-experts.

Acknowledgments

First and foremost, I am grateful for my friends, family, and partner for their unwavering support. Their encouragement, patience, and late–night conversations have brought a tremendous amount of happiness and motivation over these past four years, and this thesis is as much a product of their care as it is of my own work. My deep gratitude and respect goes to my thesis advisor, Dr. Yu Liang, for welcoming my curiosity and giving me full reins on this project. From letting me audit his machine learning class in my sophomore year to involving me in his past research, he created space for me to explore, be curious, and slowly learn how to think like a researcher. I would also like to thank Dr. Wigal for her mentorship over the past couple of years and for welcoming me with open arms into the community she has built at UTC through the Grand Challenges Scholars Program. That program has helped me become more of a well–rounded engineer. Thank you to Mr. Howell for serving on this committee and for deepening my interest in systems programming and hardware. His classes is a large part of what now motivates me to pursue embedded security in my master’s studies. Finally, I am deeply grateful to the Grand Challenges Scholars Program itself and to the cohort I entered with. Through Grand Challenges, I have learned to move from being purely problem–based to being solutions–based in a way that revolves around people: who is affected, how to build solutions for people heavily taking into account their experiences, and what it means to build technology that is actually worth having in the world.

Degree

B. S.; An honors thesis submitted to the faculty of the University of Tennessee at Chattanooga in partial fulfillment of the requirements of the degree of Bachelor of Science.

Date

5-2026

Subject

Generative artificial intelligence; Human-computer interaction; User trust

Keyword

Large Language Models (LLMs); Human–Machine Collaboration; AI User Experience; Exploratory Data Analysis (EDA); Web Applications

Discipline

Artificial Intelligence and Robotics

Document Type

Theses

Extent

ix, 56 leaves

DCMI Type

Text

Language

English

Rights

http://rightsstatements.org/vocab/InC/1.0/

License

http://creativecommons.org/licenses/by/4.0/

Download

Share

COinS