I-Corps: Translation Potential of an Artificial Intelligence-based Course Front Desk for Facilitating Student-Instructor Interactions

Texas A&M Engineering Experiment Station, College Station TX

Investigators

Abstract

The broader impact of this I-Corps project is based on the development of a novel artificial intelligence-enhanced chatbot system to provide affordable engagement and enhanced accessibility to students. By automating routine tasks, the system can free up time to focus on instructional design and teaching, alleviating the workload on instructors and teaching assistants. This innovation represents an affordable alternative for institutions (including community colleges and other non-traditional teaching institutions) that cannot afford additional teaching assistance for instructors, allowing them to focus on more complex and creative educational tasks and student engagement. The benefit of this approach is its potential to enhance teaching effectiveness, increase engagement in the learning process for large classes, and improve individualized learning, while preserving and reflecting the personalized approach of the instructor. This I-Corps project utilizes experiential learning coupled with a first-hand investigation of the industry ecosystem to assess the translation potential of the technology. The solution is based on the development of a Retrieval Augmented Generative artificial intelligence (RAG) “always available” chatbot system called ChaTA (short for Chat Teaching Assistant). The technology is an instructor-customizable interface between students and instructors, based on a novel approach to RAG prompting where a Large Language Model (LLM) is combined with an instructor supplied database of information (their notes, presentations and videos, and answers). The LLM will interpret the student questions, classify them based on instructor policy, convert them into a query of the database, and evaluate if the results answer the original question. The chatbot learns from the student ratings of the acceptability of the answers and is moderated by the instructor (to identify "hallucinations" or erroneous or nonresponsive answers) as it learns to reflect the instructor’s point of view. Unlike current approaches where a chatbot is used as a fully automated teaching assistant, the current approach is designed to help instructors by providing summary reports of student questions and acceptable answers, helping the instructor fine-tune their teaching. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.