Dynamic learning-path optimization via LSTM state modeling for international trade practice course

Abstract

International Trade Practice is a professional core course of trade related majors, whose traditional teaching content and methods can not meet the needs of high quality talents for modern foreign trade industry. The learning path can intuitively reflect the real process of learners' learning, which is significant for improving the quality of their learning experience providing personalized learning support and adapting teaching strategies. Therefore, this paper proposes a dynamic learning path recommendation model for the course International Trade Practice based on a deep learning (DL) algorithm. In this model, sequential recommendation is first enhanced by incorporating knowledge point concept coverage and difficulty characteristics into a dynamic learning environment, enabling a more comprehensive and accurate representation of the learning context. Furthermore, the model addresses adaptive curriculum planning by leveraging DL algorithms to intelligently recommend learning paths, significantly improving their adaptability and personalization. The results indicate that the model can accurately identify learners' weak knowledge points and recommend adaptive paths suitable for their learning needs based on their cognitive characteristics. The achievement of this research not only contributes to enhancing the efficiency of learning, but also provides new ideas for the application of educational technology in professional courses.