数智背景下知识图谱协同大模型赋能高职教学评价研究  

Research on Empowering Vocational Education Evaluation with Knowledge Graph Collaborative Large Model in the Context of Digital Intelligence

作　　者：张旭[1] ZHANG Xu(Tianjin College of Commerce,Tianjin 300350)

机构地区：[1]天津商务职业学院,天津300350

出　　处：《天津商务职业学院学报》2025年第1期90-96,共7页Journal of Tianjin College of Commerce

基　　金：天津市教育科学规划课题“新文科建设背景下信息技术与教育教学深度融合研究”(项目编号:CJE210260)的阶段性研究成果。

摘　　要：传统高职教学评价体系存在职业能力诊断失真、过程性评价缺失、校企数据割裂等结构性矛盾,影响教学评价的科学性与有效性。在深入推进教育数字化战略行动中,知识图谱凭借其结构化特点在高职教学评价中表现出能够精准进行学情诊断、为学生推荐个性化学习路径、基于过程性评价动态优化课程内容与教学设计的作用。但在教学实践中,因知识图谱动态更新缺乏灵活性和自动化、多模态知识图谱构建难度大、教师应用知识图谱进行教学评价的能力不足等因素影响教学评价精准性,导致数据驱动的评价机制难以落地。在人工智能技术下,知识图谱协同大模型在高职教学评价中通过智慧教学——大模型与知识图谱双驱动强化高职教学过程性评价,优化课堂教学资源;自主学习——知识图谱融合大模型构建学生画像与自适应学习系统;智能评测——知识图谱协同大模型构建高职教学评价多元化模式实现教—学—评一体化、闭环化和智能化。The traditional vocational education evaluation systems suffer from structural contradictions such as distorted diagnosis of vocational competencies,lack of process-based assessment,and fragmented school-enterprise data,which undermine the scientific validity and effectiveness of teaching evaluation.In the ongoing advancement of the digital education strategy,knowledge graphs with its structured characteristics,have demonstrated their potential in vocational education evaluation by enabling precise learning diagnostics,recommending personalized learning paths,and dynamically optimizing course content and instructional design based on process-oriented evaluation.However,in practice,challenges such as the inflexibility and lack of automation in knowledge graph updates,the difficulty of constructing multi-modal knowledge graphs,and teachers’insufficient ability to apply knowledge graphs in teaching evaluation hinder the accuracy of assessment,making data-driven evaluation mechanisms difficult to implement.Under the influence of AI technologies,the collaboration between knowledge graphs and large models enhances vocational teaching evaluation through:Smart teaching,a dual-driven approach combining large models and knowledge graphs to strengthen process-based evaluation and optimize classroom teaching resources;Autonomous learning,integrating knowledge graphs with large models to construct student profiles and adaptive learning systems;and intelligent assessment,leveraging knowledge graph-large model synergy to establish a diversified teaching evaluation model,achieving integrated,closed-loop,and intelligent"teaching-learning-evaluation"system.

关 键 词：数智 知识图谱 大模型 高职教学评价 

分 类 号：G712.4[文化科学—职业技术教育学]