Introduction

My name is Lelia Rhodes, a Master’s student in the Human Language Technology (HLT) program at the University of Arizona, set to graduate in May 2025. I also hold a Bachelor’s degree in Linguistics from the same institution, where I developed a strong foundation in the scientific study of language and an early interest in computational approaches to linguistics.

During my time in the HLT program, I’ve developed skills in Python programming, with current work expanding into JavaScript. I enjoy applying technical methods to linguistic data and have grown especially interested in Natural Language Processing (NLP) and its intersection with educational technology, accessibility, and the analysis of user-generated content. My experience includes working with libraries and tools like NLTK, spaCy, and scikit-learn, as well as pandas for efficient data manipulation and analysis—particularly when working with structured data sources like CSVs and spreadsheets.

For my internship, I’ve been working with the S.A.L.T. Center (Strategic Alternative Learning Techniques Center) at the University of Arizona. My project involves designing and developing a Google Sheets plugin that performs sentiment analysis on tutor feedback. The goal of this tool is to detect “green flags” (positive indicators) and “red flags” (potential concerns) in student-tutor interactions, enabling earlier intervention and better support for student success. This tool, designed and implemented as a solo project, builds on my understanding of sentiment scoring, regular expression processing, and user-centered design in educational contexts.

In another project based on a Kaggle competition, I tackled a stylometric classification task involving the prediction of whether two English text spans were written by the same author. I built a complete pipeline using text preprocessing, TF-IDF vectorization, cosine similarity, and a Gradient Boosting Classifier to solve this binary classification problem. This project strengthened my understanding of authorship identification techniques and demonstrated my ability to engineer features and apply supervised learning to real-world language data.

Looking ahead, I am passionate about teaching linguistics and computational methods to others. My goal is to combine my technical training and love of language to help learners—whether in the classroom or through tools I build—engage more deeply with how language works and how we can teach machines to understand it. I’m particularly excited about educational technology, the ethics of AI in language, and using HLT to support equity and accessibility.