知识图谱驱动的中医药标准数字化探索与实践*

作者：袁良志1，海佳丽1，汪 润1，邓文萍1,2，肖 勇1,2，常 凯1,2

单位：1.湖北中医药大学信息工程学院，湖北 武汉 430065； 2.湖北时珍实验室，湖北 武汉 430065

引用：引用：袁良志，海佳丽，汪润，邓文萍，肖勇，常凯.知识图谱驱动的中医药标准数字化探索与实践[J].中医药导报,2025,31(1):225-230.

DOI：10.13862/j.cn43-1446/r.2025.01.043

PDF： 下载PDF

摘要：目的：解决传统中医药标准文件管理中存在的效率低、更新慢和难以共享等问题，提出一种基于知识图谱技术的中医药标准数字化新方法，提升标准信息管理和应用的效率和准确性。方法：选取中国中医药信息学会发布的93项团体标准，通过构建中医药标准的本体模型，利用自然语言处理技术和大语言模型，对标准文件进行文本处理和数据映射，形成具有语义关联的中医药标准知识图谱，并搭建相应的管理与应用系统。结果：成功构建了中医药信息标准文件知识图谱，包括7 431个实体、9 700个关系和11 814个属性值，并将这些数据存入Neo4j图数据库。基于此开发了中医药信息标准管理与应用系统，支持标准关联图谱展示、精确检索和图谱动态更新等功能，提高了中医药标准信息的管理效率和查询效果。结论：本研究通过知识图谱技术实现了中医药标准的数字化表达和结构化管理，有效提升了标准信息的查询效率和理解度。

关键词：中医药标准；标准数字化；知识图谱；大语言模型

Abstract：

Objective: To solve the problems of low efficiency, slow updates, and difficulty in sharing in the management of traditional Chinese medicine (TCM) standard documents, and to propose a new digital method for TCM standards based on knowledge graph technology to improve the efficiency and accuracy of standard information management and application. Methods: The study selected 93 group standards released by the China Information Association of Traditional Chinese Medicine. By constructing an ontology model for TCM standards and utilizing natural language processing techniques and large language models, the standard documents were subjected to text processing and data mapping to form a semantically related knowledge graph of TCM standards, and a corresponding management and application system was built. Results: A knowledge graph for TCM information standard documents was successfully constructed, comprising 7 431 entities, 9 700 relationships, and 11 814 attribute values, and the data was stored in the Neo4j graph database. Based on this, a TCM information standards management and application system has been developed, which supports functions such as displaying standard association graphs, precise retrieval, and dynamic updating of graphs, improving the management efficiency and query effectiveness of TCM standard information. Conclusion: This study achieved the digital representation and structured management of TCM standards through knowledge graph technology, effectively enhancing the query efficiency and comprehension of standard information. 

Key words：traditional Chinese medicine standards; standard digitalization; knowledge graph; large language Model

发布时间：2025-12-01

点击量：62