Construction of an AI-Powered Interactive Digital Tutor and Its Application in Materials Science Laboratory Teaching

随着人工智能(AI)技术的迅猛发展，高等教育中的实验教学正面临着前所未有的转型机遇。文章围绕材料科学与工程专业核心课程《材料科学基础实验》的教学改革，构建了一个融合知识图谱(Knowledge Graph, KG)、大语言模型(Large Language Model, LLM)与智能反馈机制的AI交互式数字教师系统。该系统在知识重构、智能问答、个性推荐与资源调度等方面实现了教学智能化，显著提升了学生的学习主动性、知识迁移能力与操作规范意识。通过双模型融合保障系统稳定与数据安全，并以开放式界面优化用户体验，实现对学生认知路径的动态感知与精准支持。教学实践表明，该系统不仅有效缓解了教学资源紧张与个性化指导难题，也推动了教师角色转型与教学范式变革。本文的探索为人工智能技术在高等教育实验教学中的应用提供了实践样本与理论启示。With the rapid advancement of artificial intelligence (AI), experimental teaching in higher education is undergoing a significant transformation. This study focuses on reforming the core course Fundamentals of Materials Science Laboratory by developing an AI-powered interactive digital tutor system that integrates Knowledge Graphs (KG), Large Language Models (LLM), and Intelligent Feedback Mechanisms. The system promotes intelligent teaching through semantic knowledge reconstruction, real-time Q&A, personalized recommendation, and intelligent resource scheduling, thereby enhancing students’ engagement, knowledge transfer ability, and operational competence. A dual-model architecture ensures both performance and data security, while an interactive interface enables contextual understanding and personalized support. Teaching practice demonstrates that the system effectively addresses challenges such as limited resources and lack of individualized guidance, facilitates the shift in teachers’ roles, and drives pedagogical innovation. This work offers both practical insights and theoretical references for AI-enabled experimental education in universities.