Bridging LMS and Generative AI: Dynamic Course Content Integration (DCCI) for Connecting LLMs to Course Content -- The Ask ME Assistant
Project Overview
This document explores the integration of Generative AI, specifically Large Language Models (LLMs), into Learning Management Systems (LMS) to improve educational experiences. It introduces the Dynamic Course Content Integration (DCCI) mechanism, which enables LLMs to deliver accurate and context-aware responses to student queries by linking them directly to course content. The implementation of the Ask ME Assistant Bot showcases significant user satisfaction and engagement, illustrating the potential of AI to enhance learning. However, the document also addresses challenges such as the risk of over-reliance on AI technologies, concerns regarding the accuracy of AI-generated information, and ethical considerations in the educational context. Overall, the findings suggest that while Generative AI has the potential to enrich educational experiences, careful attention must be given to its limitations and the implications for teaching and learning.
Key Applications
Ask ME Assistant Bot
Context: Higher education, specifically for first-year programming students at Eötvös Loránd University.
Implementation: The bot was integrated into the Canvas LMS using the DCCI mechanism, allowing it to dynamically retrieve course content and provide context-aware responses.
Outcomes: High user satisfaction (average rating of 4.614/5), improved engagement, reduced platform switching, and support for self-directed learning.
Challenges: Issues with hallucination (inaccurate responses) and over-reliance on AI, concerns about accuracy, plagiarism, and reduced student-teacher interaction.
Implementation Barriers
Technical
Hallucination where LLMs generate inaccurate or misleading information.
Proposed Solutions: Implementing the DCCI mechanism to ensure responses are anchored in dynamically retrieved course materials.
Ethical
Concerns regarding over-reliance on AI and potential academic dishonesty.
Proposed Solutions: Establishing ethical guidelines for AI usage in education and promoting human-AI collaboration.
Integration
Challenges in integrating AI tools with existing educational practices and ensuring they do not replace human interactions.
Proposed Solutions: Developing a pedagogical framework that prioritizes collaboration between AI tools and educators.
Project Team
Kovan Mzwri
Researcher
Márta Turcsányi-Szabo
Researcher
Contact Information
For information about the paper, please contact the authors.
Authors: Kovan Mzwri, Márta Turcsányi-Szabo
Source Publication: View Original PaperLink opens in a new window
Project Contact: Dr. Jianhua Yang
LLM Model Version: gpt-4o-mini-2024-07-18
Analysis Provider: Openai