中医经方知识图谱“图搜索模式”设计研究

Study on Design of Graph Search Pattern of Knowledge Graph of TCM Classic Prescriptions

目的设计中医经方知识图谱及图搜索模式,为经方知识的浏览、检索及运用提供新的设计思路。方法采用领域本体七步法,以中医经方为研究对象进行知识图谱模式层和数据层设计。运用Neo4j中的Cypher语言设计一种模拟人联想思维进行检索的经方知识检索框架。结果设计的中医经方知识图谱结构,概念包括方剂、治疗、中药、禁忌证、功用等,概念与概念间关系有方剂→组成部分→中药、方剂→禁忌→证等。模拟人联想思维进行检索的经方知识检索框架包含随证选方、方剂使用禁忌与煎服方法等功能;利用拓扑有向图认知结构模拟"方证对应"的过程;可进行单个医案的经方用药分析及以方测证等功能的智能推理。结论本研究设计的中医经方知识图谱及图搜索模式,可模拟人联想思维进行经方数据挖掘及分析。

Objective To design the knowledge graph of TCM classic prescriptions and its corresponding search mode;To provide a new design idea for browsing, retrieving and using TCM knowledge for case analysis.Methods Taking the TCM classic prescription as the research object, the domain ontology seven-step method was used to design knowledge map pattern layer and data layer. Cypher language of the Neo4j map database was used to design a kind of classic prescriptions retrieval framework by simulating human associative thinking. Results The designed structure of knowledge graph of TCM classic prescription: concepts included prescriptions, treatment,Chinese materia medica, contraindications and functions. The relationship between concepts was prescription →component → Chinese materia medica, prescription → taboo → syndrome. The functional knowledge retrieval framework through simulating human associative thinking included functions such as the selection of prescriptions,the use taboos of prescriptions and the method of decoction;the process of simulating the "prescription and syndrome correspondence" by using the topological directed graph cognitive structure;the intelligent reasoning of the functions of the individual medical treatment, such as prescription analysis and prescription testing. Conclusion Graph search pattern and knowledge graph of TCM classic prescriptions designed in this study can simulate human associative thinking to conduct data mining and analysis for classic prescriptions.