Enhancing STEM Education with Artificial Intelligence: A Virtual Teaching Assistant to Engage in Effective Pedagogical Methods

This project aims to serve the national interest by advancing STEM education through the development and evaluation of artificial intelligence (AI) driven virtual teaching assistants (VTAs) that support an on-demand, personalized learning environment for students. Current advances in generative AI language models create an unprecedented opportunity to improve STEM education. This project aims to combine state-of-the-art in AI language models with effective pedagogical techniques, such as inquiry-based learning and the Socratic method. The project emphasizes readily available access to academic support through scalable, open-source technology that addresses challenges posed by large class sizes and limited instructor availability. The significance of this work lies in its potential to improve student outcomes, promote reflective learning, and enhance learning environments. The project's goals include (a) the creation of a generalizable Large Language Model (LLM) framework for constructing conversational AI agents, i.e., VTAs, optimized for STEM instruction, (b) the deployment and optimization of VTAs in thermo-fluids courses, and (c) the evaluation of the pedagogical effectiveness using both quantitative metrics and student feedback. The project plans to assess the accuracy of VTA responses, measure student learning outcomes, and analyze conversation logs between students and the VTA to extract insights about student understanding and sentiment. Findings will support the broader implementation of AI-enhanced teaching tools in STEM education and inform the efficacy of such tools when compared to human tutors. The NSF IUSE: EDU Program supports research and development projects to improve the effectiveness of STEM education for all students. Through the Engaged Student Learning track, the program supports the creation, exploration, and implementation of promising practices and tools. 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.