An Exploration of the Logical Data and Knowledge Graph of Wang Xugao's Case Records Based on Knowledge Element Indexing
-
摘要:
目的 采用知识元理论的信息技术梳理、分析中医古籍内容,构建知识图谱探析隐含的逻辑关系以发现新知识。 方法 以《王旭高医案》为例,在基于知识元理论与技术深度标引的基础上,首先利用MS SQL Server数据库将标引数据读取为逻辑数据并初步分析;其次以基于neo4j数据库构建的中医古籍知识图谱技术呈现出显性知识,同时探析其深层的逻辑推理关系,进一步发现隐性知识。 结果 《王旭高医案》共有知识体787个,知识元5 153个,语义类型共有1 149个,语义关联共有510个。分析逻辑数据和机构化知识图谱可知,虚劳的知识元与语义关联最多,其中与肝脏、脾胃的语义关联最多。 结论 《王旭高医案》整体来看特点在于重视对各类疾病证候表现的描述以及病因病机的分析, 王氏诊治虚劳的经验较为丰富,从肝入手诊疗虚劳,尤其重视肝脾同病的病机。 Abstract: OBJECTIVE To sort out and analyze the contents of Chinese medical classics by using the information technology of knowledge element theory, and to find new knowledge through exploring the implied logical relationships and constructing a knowledge graph.METHODS Through depth indexing of Wang Xugao's Case Records based on knowledge element theory and technology, we firstly read out indexing data as logical data and carried out preliminary analysis with MS SQL Server database. Secondly, we presented the explicit knowledge by using the knowledge graph technology of Chinese medical classics based on neo4j database, and explored the underlying logical relationships to further discover the implicit knowledge.RESULTS There were 787 knowledge bodies, 5 153 knowledge elements, 1 149 semantic types, and 510 semantic associations in Wang Xugao's Case Records. By analyzing the logical data and the structured knowledge graph, it can be seen that the knowledge elements and semantic associations of deficiency-consumption in the book are the most numerous, among which the semantic associations with liver, spleen and stomach are the most numerous.CONCLUSION It is concluded that Wang Xugao's Case Records is characterized by the emphasis on the description of syndromes and symptoms of various diseases and the analysis of relevant etiologies and pathogenesis. Besides, Wang had rich experience in the diagnosis and treatment of deficiency-consumption. His treatment of deficiency-consumption starts with the liver, with particular attention to the pathogenesis of the disease involving both the liver and spleen. -
表 1 知识元统计
知识元名称 数量 占比/% 证候表现 922 17.9 方剂 798 15.5 医案名 785 15.2 病因病机 558 10.8 姓名 530 10.3 治则治法 519 10.1 评按 205 4.0 脉象 202 3.9 舌象 130 2.5 预后 119 2.3 辨证 100 1.9 其他 285 5.5 表 2 语义类型统计
语义类型 数量 占比/% 病因病机 421 36.6 证候表现 391 34.0 病证 119 10.4 脉象 95 8.3 舌象 42 3.7 治法 41 3.6 药物 25 2.2 方剂 13 1.1 治则 2 0.2 表 3 语义关联统计
语义关联 数量 占比/% 证因关系 403 79.0 证象关系 64 12.5 辨治关系 40 7.8 证治关系 2 0.4 方药关系 1 0.2 -
[1] 索传军, 盖双双. 知识元的内涵、结构与描述模型研究[J]. 中国图书馆学报, 2018, 44(4): 54-72. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGTS201804004.htm [2] 韦昌法, 晏峻峰. 从知识表示与推理方法探讨中医数字辨证发展[J]. 中华中医药杂志, 2019, 34(10): 4472-4473. https://www.cnki.com.cn/Article/CJFDTOTAL-BXYY201910004.htm [3] 张洁瑜. 王旭高医案研究[D]. 北京: 北京中医药大学, 2019. [4] 严季澜, 陈仁寿. 中医文献学[M]. 北京: 人民卫生出版社, 2016: 131-132. [5] 高国伟, 王亚杰, 李永先. 我国知识元研究综述[J]. 情报科学, 2016, 34(2): 161-165. https://www.cnki.com.cn/Article/CJFDTOTAL-QBKX201602032.htm [6] 王勇, 王凤兰. 《黄帝内经》知识表示与标引研究[J]. 山东中医药大学学报, 2020, 44(5): 585-590. https://www.cnki.com.cn/Article/CJFDTOTAL-SDYX202005027.htm [7] 朱玲, 于彤, 杨峰. 基于关键动词的中医古籍概念实体间语义关系发现研究[J]. 中国数字医学, 2016, 11(5): 73-75. doi: 10.3969/j.issn.1673-7571.2016.05.023 [8] 王清华. 《金匮·虚劳病》调补脾肾法浅析[J]. 湖北中医杂志, 1998, 20(6): 11-12. https://www.cnki.com.cn/Article/CJFDTOTAL-HBZZ199806004.htm [9] 戴梦, 刘杰, 盛昭园. 中国历代医家对虚劳病的认识之回顾与分析[J]. 环球中医药, 2021, 14(4): 627-628. doi: 10.3969/j.issn.1674-1749.2021.04.016 [10] 王旭高. 西溪书屋夜话录[M]. 北京: 人民军医出版社, 2012: 1.