Authors:

Ekagrah Kumar

Date Created:

Not specified

Course Title:

Professor:

Not specified

About Paper:

This research addresses the growing need for AI-driven educational support in complex, interdisciplinary fields such as biological engineering. An AI tutor was developed and tested for Purdue University's ABE 20100 (Material and Energy Balances) course, aiming to enhance student understanding, develop engineering problem-solving skills, and foster competence as future engineers. The methodology utilizes Large Language Models (LLMs) with Retrieval-Augmented Generation (RAG), allowing the tutor to generate contextually relevant and accurate responses directly from provided course materials, including the syllabus, textbooks, and class slideshows. Prompt Engineering was used to define the tutor's role as a patient, knowledgeable, and adaptive educator that uses the Socratic method for teaching. The system employs evidence-based and Personalized Adaptive Learning (PAL) strategies, such as active recall, spaced repetition with self-assessment, and immediate, customized exercise feedback, tailoring the learning experience to individual student progress and identifying difficulties. Preliminary testing using text-based quizzes indicates the tutor's effectiveness in promoting student engagement and improving understanding through interactive exercises and tailored guidance. A key finding is that AI tutors can significantly complement traditional instruction, offering scalable, adaptive, and personalized learning experiences crucial for mastering interdisciplinary engineering concepts. Preliminary studies demonstrated AI's potential to reduce learning time and enhance efficiency in educational settings. However, challenges persist with accurately handling complex numerical and analytical problems, which future developments must address. More complete evaluations will be conducted on learning outcomes in ABE 20100 when the course is next offered. Keywords: Biological Engineering; Engineering Education; Artificial Intelligence; Large Language Models; Intelligent Tutoring Systems † Presenting Undergrad Author; ‡ Contributing Undergrad Author; * Undergrad Acknowledgment

Source:

Purdue University / 2025

Topics:

No topics listed

Co-authors:

Ekagrah Kumar