Scholarlink

Software modeling education often lacks immediate and personalized feedback, making it challenging for novice learners to comprehend modeling principles. This paper continues our ongoing work on developing an AI-driven feedback mechanism to support UML diagram construction. Building on previous efforts, we improved the system by enhancing the RAG-LLM component within the existing UML Miner plugin. The system analyzes students’ modeling behavior and provides real-time, personalized guidance to support learning and improve modeling skills.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A RAG-Enhanced AI Feedback for UML Education

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Pasquale Ardimento,
  • Mario Luca Bernardi,
  • Marta Cimitile,
  • Michele Scalera

摘要

Software modeling education often lacks immediate and personalized feedback, making it challenging for novice learners to comprehend modeling principles. This paper continues our ongoing work on developing an AI-driven feedback mechanism to support UML diagram construction. Building on previous efforts, we improved the system by enhancing the RAG-LLM component within the existing UML Miner plugin. The system analyzes students’ modeling behavior and provides real-time, personalized guidance to support learning and improve modeling skills.