<![CDATA[The rapid advancement of artificial intelligence, particularly large language models (LLMs), has significantly transformed the landscape of digital learning environments. In programming education, students often face difficulties such as limited instructor availability, delayed feedback, and insufficient personalized guidance during the learning process. Intelligent Tutoring Systems (ITS) have been widely proposed as a solution to provide adaptive and individualized learning support. However, traditional ITS architectures often rely on predefined rule-based models that limit their scalability and contextual understanding. This study proposes a large language model-based intelligent tutoring system designed to enhance programming education through adaptive learning support, automated feedback, and natural language interaction between students and the system. The proposed framework integrates LLM capabilities with a tutoring architecture that supports real-time code explanation, debugging assistance, and concept clarification tailored to individual learner needs. The system leverages prompt engineering and retrieval mechanisms to improve response relevance and pedagogical effectiveness. The results demonstrate that integrating LLM technologies into tutoring systems can improve students’ learning engagement, programming performance, and problem-solving abilities. Furthermore, the proposed approach enables scalable educational assistance that can support learners in environments with limited teaching resources. The findings suggest that LLM-based tutoring systems have strong potential to become an effective solution for personalized programming education in modern digital learning ecosystems.]]>
Copyrights © 2025
