• 中国核心期刊(遴选)数据库收录期刊
  • 中文科技期刊数据库收录期刊
  • 中国期刊全文数据库收录期刊
  • 中国学术期刊综合评价数据库统计源期刊等

中国药物评价 ›› 2026, Vol. 43 ›› Issue (1): 24-24-31.doi: 10.2095-3593.2026.030004

• 药物研究 • 上一篇    下一篇

基于知识图谱的大理民族民间痹证单验方用药规律研究

马铮然1, 黄子易1, 王肖盈1, 杨文龙2, 朱志伟2*   

  1. 1.浙江中医药大学第三临床医学院, 浙江 杭州 310053;
    2.云南省大理白族自治州中医医院, 云南 大理 671000
  • 收稿日期:2025-11-11 修回日期:2026-01-08 接受日期:2026-04-01 出版日期:2026-02-28 发布日期:2026-04-01
  • 基金资助:
    云南省科技厅中医联合专项-面上项目(20202301AZ070001-115);云南省科技厅中医联合专项-青年项目(202301AZ070001-178);云南省科技厅中医联合专项-青年项目(202401AZ070001-141);云南中医药大学校院联合基金项目-重点项目(XYLH2024023)

A Knowledge Graph-Based Study on the Medication Rules of Dali Ethnic Folk Empirical Prescriptions for Bi Syndrome

 MA Zhengran1, HUANG Ziyi1, WANG Xiaoying1, YANG Wenlong2, ZHU Zhiwei2*   

  1. 1.The Third Clinical Medical School of Zhejiang Chinese Medical University, Zhejiang Hangzhou 310053, China;
    2.Dali Bai Autonomous Prefecture Chinese Medicine Hospital, Yunnan Dali 671000, China
  • Received:2025-11-11 Revised:2026-01-08 Accepted:2026-04-01 Online:2026-02-28 Published:2026-04-01

摘要: 目的:运用知识图谱技术构建大理民族民间痹证单验方知识图谱,挖掘其方药配伍规律,探索知识图谱在民族民间医药研究中的应用价值。方法:通过查询《中国分省医籍考》、地方县志等资料,整理大理地区的医药典籍,汇总民间痹证相关单验方。采用本体构建技术进行知识建模,并依据知识抽取、知识表示、知识存储及可视化等步骤实现知识图谱构建,于Neo4j图数据库中呈现大理民族民间痹证单验方知识图谱。在此基础上,对以“方-药”为核心的多关系知识网络进行图算法与社区发现分析。结果:共构建包含773个实体、1172条关系的知识图谱,形成以“方-药”为核心的多关系知识网络。分析发现3个核心方剂,其交集药物“杜仲、独活、秦艽、羌活”构成核心药物组合,其中续断为最核心的药物,当归次之,并发现以青阳参、千年健、石椒草为代表的关键特色民族医药,揭示其潜在临床应用价值。结论:本研究系统地呈现了大理民族民间痹证单验方的“方-药”知识网络,揭示其以祛邪扶正、补益肝肾为核心的治则特点,凸显地域用药独特性。通过将知识图谱与图论算法应用于小样本、稀疏性民间方药数据,为民族民间单验方的科学化与规范化研究提供了可复用的方法论框架。

关键词: font-size:medium, ">民族民间单验方;痹证;知识图谱;数据挖掘

Abstract: Objective: To construct a knowledge graph of Dali ethnic folk empirical prescriptions for Bi syndrome using knowledge graph techniques, to identify formula-herb compatibility patterns, and to explore the application value of knowledge graphs in ethnic folk medicine research. Methods: Medical literature from the Dali region was systematically collected through sources such as Chinese Medical Works in Different Provinces, local gazetteers, and folk empirical prescriptions for Bi syndrome were collected. Ontology-based knowledge modeling was employed, followed by knowledge extraction, representation, storage, and visualization to construct the knowledge graph, which was implemented in the Neo4j graph database. On this basis, graph algorithms and community detection methods were applied to analyze the multi-relational knowledge network centered on the “formula-herb” relationship. Results: A knowledge graph comprising 773 entities and 1172 relationships was constructed, forming a multi-relational knowledge network centered on the “formula-herb” structure. Three core formulas were identified, and their shared herbs (Duzhong, Duhuo, Qinjiao and Qianghuo) constituted the core herbal combination. Among them, Xuduan was identified as the most central herb, followed by Danggui. In addition, key characteristic ethnic medicinal herbs represented by Qingyangshen, Qiannianjian, and Shijiaocao were identified, revealing their potential clinical application value. Conclusion: This study systematically presents the “formula-herb” knowledge network of Dali ethnic folk empirical prescriptions for Bi syndrome, revealing therapeutic principles characterized by dispelling pathogenic factors while supporting healthy qi and tonifying the liver and kidney, and highlighting distinct regional medicinal characteristics. By applying knowledge graph techniques and graph-theoretical algorithms to small-sample and highly sparse folk prescription data, this study provides a reusable methodological framework for the scientific and standardized research on ethnic folk empirical prescriptions.

Key words: font-size:medium, "> , Ethnic folk empirical prescriptions; Bi syndrome; Knowledge graph; Data mining

中图分类号: