Designing Learner-Centered Collaborative Learning by Incorporating AI-Based Teacher/Learner Agents with a Cognitive Model
DOI:
https://doi.org/10.58459/icce.2024.4821Abstract
This paper presents collaborative concept-mapping tutor (CoCot ver.2), a collaborative learning support system that integrates concept maps with a conversational agent. CoCot ver.2 features two agents: a teacher agent and a student agent. The teacher agent acts as a human instructor, engaging in conversations with learners, aiding their metacognition, and summarizing the discussion content. The student agent learns from the learners' concept map creation and generates their own concept map knowledge. These agents are developed using (1) a cognitive architecture (ACT-R) for knowledge generation for the agents' concept map and (2) GPT 3.5 for part of the language processing for agent-based feedback.
Downloads
Download data is not yet available.
Downloads
Published
2024-11-25
Conference Proceedings Volume
Section
Articles
How to Cite
Designing Learner-Centered Collaborative Learning by Incorporating AI-Based Teacher/Learner Agents with a Cognitive Model. (2024). International Conference on Computers in Education. https://doi.org/10.58459/icce.2024.4821