知识图谱与AI大模型协同驱动的材料物理学教学改革实践
Teaching Reform Practice of Material Physics Enabled by Knowledge Graph and Large-Scale AI Models
DOI: 10.12677/ces.2026.1411885, PDF,    科研立项经费支持
作者: 秦 蒙*, 王毅轩, 杨俊涛, 童 锐, 周望怀, 周 峒:湖北汽车工业学院新能源学院,湖北 十堰
关键词: 材料物理学;知识图谱;AI大模型;教学改革;Materials Physics; Knowledge Graph; AI Large Models; Teaching Reform
摘要: 材料物理学作为材料科学与工程专业的核心基础课程,具有学科交叉性强、知识体系庞杂、概念抽象等特点,传统教学模式面临知识碎片化、理论与应用脱节、学情反馈滞后等突出问题。本文以“知识图谱为中枢、AI大模型为接口”的协同驱动理念为核心,提出并实践了一套材料物理学课程教学改革方案。通过系统梳理课程知识体系,构建涵盖晶体结构、缺陷物理、相变与扩散等核心模块的结构化知识图谱,实现知识点间关联关系的显性化与可视化;在此基础上,依托大语言模型开发课程专属AI智能助教系统,实现智能答疑、个性化学习路径推荐与学情自动分析三大功能。教学实践表明,该模式有效提升了学生对抽象概念的理解深度和知识体系的整体把握能力,同时为教师提供了精准的教学决策支持,为材料类基础课程的智能化教学改革提供了可行路径。
Abstract: Materials Physics, a core foundational course for the Materials Science and Engineering major, is characterized by strong interdisciplinary integration, a complex and extensive knowledge system, and highly abstract concepts. Traditional teaching models face prominent challenges including fragmented knowledge delivery, a disconnect between theory and application, and delayed feedback on students’ learning progress. Centering on the collaborative-driving philosophy of “knowledge graph as the hub, AI large model as the interface,” this paper proposes and implements a teaching reform scheme for the Materials Physics course. By systematically reorganizing the course’s knowledge framework, a structured knowledge graph was constructed covering core modules such as crystal structure, defect physics, and phase transformation and diffusion, thereby making the relationships among knowledge points explicit and visually accessible. On this basis, a course-specific AI intelligent teaching-assistant system was developed leveraging large language models, which delivers three major functions: intelligent Q&A, personalized learning-path recommendation, and automatic analysis of students’ learning status. Teaching practice demonstrates that this model effectively deepens students’ understanding of abstract concepts and strengthens their holistic command of the knowledge system, while also providing instructors with precise teaching-decision support, offering a feasible pathway for the intelligent reform of foundational courses in materials-related disciplines.
文章引用:秦蒙, 王毅轩, 杨俊涛, 童锐, 周望怀, 周峒. 知识图谱与AI大模型协同驱动的材料物理学教学改革实践[J]. 创新教育研究, 2026, 14(11): 257-266. https://doi.org/10.12677/ces.2026.1411885

参考文献

[1] 高卫红, 傅宇东, 孙斌, 等. 新工科背景下材料物理学课程教学创新改革的思考与实践[J]. 高教学刊, 2026, 12(6): 76-79, 84.
[2] 刘高斌, 王森, 卢艳青, 等. 材料专业“固体物理”课程混合式教学改革探讨[J]. 教育教学论坛, 2026(25): 126-130.
[3] 王戈明, 刘繁, 匡双阳, 赵洪阳, 邵晨冬. 材料物理专业实习教学模式的创新与实践[J]. 创新教育研究, 2025, 13(6): 599-605.
[4] 王鹏, 张居敏. 物理与数据双驱动的材料力学AI教学实践[J]. 教育进展, 2026, 16(9): 535-543.
[5] 颉雨佳, 刘叶, 朱裔荣, 陈文昊. 新工科背景下材料固体物理课程教学改革的探索和实践[J]. 创新教育研究, 2025, 13(12): 53-60.
[6] 唐剑锋, 王放, 李冠男, 等. 产出导向的课程设计助力研究生综合能力培养——以材料物理学课程为例[J]. 西南师范大学学报(自然科学版), 2022, 47(8): 117-124.