Business

Anita Carson

Professor, Operations & Technology Management

Contact me

Options

Volunteer Basis, Independent Funding Available, Potential for UROP Funding

Overview

NLP-Oriented RA Position

Have you ever seen a generative AI-powered tool (e.g., ChatGPT) produce an obvious or surprising error? As AI-powered tools rapidly expand across industries, especially in healthcare, understanding why these errors occur and how to avoid them is becoming increasingly important.
We are seeking a technically skilled and intellectually curious undergraduate student to support an interdisciplinary research project focused on natural language processing (NLP), machine learning, and automated error detection in AI-generated medical notes.

About the project:
Our research examines the accuracy and reliability of AI-generated medical documentation. Our research focuses on detecting potential errors in AI-generated medical notes. The project involves:
● Training and evaluating NLP models to detect potential AI errors
● Working with real-world unstructured medical data
● Contributing to the development of an automated error-detection pipeline for medical documentation

Learning & Professional Development Opportunities:
● Work closely with a supportive research team at the Questrom School of Business
● Build practical skills in NLP pipelines, model evaluation, and error analysis with a real clinical dataset.
● Strengthen experience relevant to careers in ML engineering, data science, healthcare analytics, and AI research

Preferred Qualifications:
Students from Computer Science, Data Science, Medical Technology, Business, or related fields are encouraged to apply. The ideal candidate will have:
● Strong proficiency and experience with NLP libraries (spaCy, Hugging Face, scikit-learn) in Python or other similar programming languages.
● Work comfortably with GitHub, including pull requests, push requests, branch management, and code reviews
● Preprocess and analyze large sets of clinical transcripts, and evaluate model performance using metrics (e.g., F1, accuracy, precision/recall)
● Ability to work independently and reliably meet deadlines
● Strong writing and communication skills

Time Commitment:
● 5 hours per week, flexible scheduling
● At least one semester commitment, with the option to continue into the summer or the next academic year
● Potentially paid position (rate aligned with BU undergraduate research assistant pay)

How to Apply:
Please send the following to Ruozhu Wang (ruozhuw@bu.edu) with the subject line “Undergraduate Research Assistant Application – [Your Name]”:
1. A brief cover letter about your interest in the project
2. Your résumé