Generation of personalized learning paths in the digitalization of education: a study on dynamic optimization based on large language models

Abstract

This paper presents a personalized learning path generation method based on a dynamic optimization mechanism, which addresses the issues of student behavior change response and path accuracy. By combining task decomposition and knowledge point association techniques, and through the construction of a mapping relationship between learning content and goals, the learning path recommendation is achieved. The memory-based collaborative filtering algorithm and the automatic tag generation mechanism are adopted to enhance the personalization and targeting of the path recommendation. At the same time, an adaptive improvement strategy is proposed, which dynamically adjusts the path based on the real-time learning feedback and behavior data of students. The non-synchronous training and distributed computing methods are utilized to improve the processing capacity and response speed of the system. The experimental results show that this method outperforms traditional recommendation algorithms in terms of accuracy and timeliness, providing efficient and scalable technical support for personalized learning.