Search

Design and Teaching Practice of an Intelligent Learning Path Planning System Based on Large Language Models and Multi-Agent Collaboration

Author(s)

Wei Liu*, Fang Li, Lingling Tan, Yafan Chen, Lianpeng Li

Affiliation(s)

School of Automation, Beijing Information Science and Technology University, Beijing, China *Corresponding Author

Abstract

To improve personalized teaching quality and autonomous learning outcomes, and to address the common problems in traditional online education, including uniform learning paths, fragmented knowledge systems, and rapid forgetting after learning, this paper takes the design and implementation of an intelligent learning path planning system as a case of teaching reform. The system integrates large language models, knowledge graphs, and multi-agent technologies, and builds core functional modules for intelligent path generation, multi-channel retrieval enhancement, adaptive replanning, and review scheduling. The system supports a shift in teaching practice from a course-centered model to a learner-centered model. The teaching practice results show that the proposed system, which is based on advanced artificial intelligence technologies, effectively reduces students’ cognitive load, stimulates their interest in autonomous learning, and improves the teaching effectiveness and adaptive intervention capability of intelligent education systems.

Keywords

Learning Path Planning; Multi-Agent; Large Language Model; Knowledge Graph; Adaptive Learning; Teaching Reform

References

[1] Gu Xiaoqing, Liu Tong. Research on Intelligent Adaptive Learning in the Era of Large Models: Progress, Cases and Prospects. The Chinese Journal of ICT in Education, 2024, 30(05):55-66. [2] Wu Di, Li Huan, Chen Xu. Analysis of the Educational Application Impact of General Artificial Intelligence Large Models. Open Education Research, 2023, 29(02):19-25. [3] Li Jiahou. The Impact of Generative Artificial Intelligence on Curriculum, Teaching Materials and Teaching Methods. Curriculum, Teaching Material and Method, 2024, 44(02):14-21. [4] Chen Xiangdong, Wang Juan, Pan Yaya, et al. Prospects for Educational Applications of Large Models: A Technology Foresight-Based Approach. e-Education Research, 2023, 44(09):5-11. [5] Zhu Zhiting, Dai Ling, Zhao Xiaowei, et al. Evolution and Educational Applications of Subject Knowledge Graphs Driven by Multimodal Large Models. Modern Educational Technology, 2023, 33(12):76-88. [6] Yi Kaiyu, Han Xibin. From Blended Teaching to Human-AI Collaborative Teaching: A New Form of Teaching under the Technological Transformation of Generative Artificial Intelligence. China Distance Education, 2024(04):85-98. [7] Jiang Qiang, Zhao Wei, Wang Pengjiao, et al. Research on Generative Recommendation Strategies for Accurate Personalized Learning Paths Based on Learner Profiles. China Educational Technology, 2019(05):84-91. [8] Mou Zhijia, Wu Fayuan. Research on Personalized Learning Models and Support Mechanisms Based on Knowledge Graphs. China Educational Technology, 2022(05):75-81. [9] Zhang Xuejun, Dong Xiaoxiao. Generation Mechanism and Application Exploration of Personalized Learning Paths for Smart Classrooms. e-Education Research, 2021, 42(04):69-75. [10] Yang Wenxia, Wang Weihua, He Lang, et al. Research and Practice of Knowledge Graph-Empowered Smart Education: A Case Study of the Linear Algebra Course at Wuhan University of Technology. Research in Higher Education of Engineering, 2023(06):111-117.