【摘要】 大模型的兴起为知识图谱平台的智能化跃迁提供了新契机，但关于大模型何以深度赋能知识图谱平台智能化升级的理论机理与实践路径尚不明确。基于此，文章对11个主流知识图谱平台的构建方法、生成方式与功能表现进行了内容编码，识别出当前平台在知识抽取精准度、知识融合深度、知识推理能力方面存在的局限。针对上述痛点问题，文章探析了大模型赋能知识图谱平台服务增强的方向。在此基础上，文章从环境建设、能力支撑、场景应用三个层面，构建了大模型赋能的高校知识图谱平台智能化升级框架。依托此框架，文章从教师、学生、管理者等不同教育主体的视角，提出了知识图谱平台与高校教育教学深度融合的落地路径，以期为大模型时代知识图谱的智能化构建与高质量应用提供参考，助力高等教育在智能时代的内涵式发展。更多还原

【Abstract】 The emergence of large language models provides a new opportunity for the intelligent leap of knowledge graph platform. However, the theoretical mechanism and practical paths for how large language models can deeply empower the intelligent upgrade of knowledge graph platform remain unclear. Based on this, the paper conducted content coding on the construction methods, generation methods, and functional performances of 11 mainstream knowledge graph platforms, identifying the limitations of current platforms in terms of the accuracy of knowledge extraction, the depth of knowledge integration, and the ability of knowledge reasoning. In response to the aforementioned pain points, this paper explored the directions for enhancing knowledge graph platform services empowered by large language models. On this basis, the paper constructed an intelligent upgrade framework of university knowledge graph platform empowered by large language models from three levels of environmental construction, capability support, and scenario application. Based on this framework, the paper proposed the implementation paths for the deep integration of knowledge graph platform with university education and teaching from the perspectives of different educational subjects such as teachers, students, and administrators, with the aim of providing references for the intelligent construction and high-quality application of knowledge graphs in the era of large language models and helping higher education achieve connotative development in the intelligent era.更多还原