Provide personalized programming learning for individuals based on large language models
Chong Li, Xin Lee, Xiaoliang Wu · Alexandria Engineering Journal · 2025
General programming learning often struggles to keep up with the rapid evolution of languages, frameworks, and applications. Moreover, most learning sources follow the same pattern, which may not suit everyone, especially beginners who require a more tailored approach. To address this, this paper proposes an innovative approach that leverages dynamic and continuously updated programming learning resources by combining network resources and large language models. We have designed a programming language learning process that includes generating personalized learning materials, providing real-time feedback, and planning subsequent learning steps. By integrating code analysis and generation technologies with vast internet resources, our system delivers a personalized language learning experience. The key innovations include the process of constructing a comprehensive code knowledge base for various programming languages and implementing progress-based learning tracking and knowledge reinforcement. To evaluate the effectiveness of this method, we conducted a controlled experiment with 20 students learning a programming language. The results demonstrate that our approach significantly accelerates students' learning efficiency, providing valuable insights and methods for programming language education.