体现辨证论治差异的不孕症知识图谱构建方法研究

doi:10.19678/j.issn.1000-3428.0064118

摘要/Abstract

摘要： 不同医家在辨证论治、辨证结果及用药习惯上存在差异，目前已有的知识图谱只对辨证论治知识进行关联与表达，不能直观地体现辨证论治差异。以6位医家的不孕症知识图谱为基础，通过引入“医家”与“关联强度”两种关系属性，构建一个整合多位医家辨证论治知识并直观体现差异性的知识图谱。在PageRank算法对不孕症数据进行预处理的基础上，通过对辨证论治过程自定义逻辑规则，利用概率软逻辑推理关联强度。建立基于关系图注意力网络的知识融合模型，考虑关系对实体含义表达的影响，通过关系图注意力网络层加权传播与聚合邻居实体信息，得到更丰富的实体向量表示从而实现不孕症知识融合。实验结果表明，该模型具有较好的融合能力，在朱南孙-钱伯煊数据集上Hit@1、Hit@10、Hit@30、MeanRank均取得较好结果，分别达到45.63%、60.85%、91.55%、0.564。通过构建体现辨证论治差异的不孕症知识图谱，可以系统地建立多位医家辨证论治知识之间的关联，对中医个性化知识的传承与发展具有重要意义。

关键词: 知识图谱, 辨证论治, 不孕症, 概率软逻辑, 图神经网络

Abstract: Variations can be observed between the dialectical results and medication treatment of doctors because doctors differentiate and treat syndromes differently.At present, existing knowledge graphs can only relate and express formulated dialectical knowledge but cannot intuitively reflect the differences in syndrome differentiation and treatment between doctors.In this study, a knowledge graph representing information on infertility integrating the knowledge on syndrome differentiation and treatment of six doctors is constructed by introducing relationship attributes indicating individual doctors and the strength of association between knowledge items.First, custom logic rules are defined for the process of syndrome differentiation and treatment;association strength is deduced by applying probabilistic soft logic based on the data describing pretreatment examinations for infertility using the PageRank algorithm.Second, this paper presents a knowledge fusion model based on a relational graph attention network;it is designed to consider the influence of relationships on expressions of the meaning of entities and to obtain a richer entity vector representation to fuse infertility knowledge by employing weighted propagation and aggregation of neighboring entity information through a relational graph attention network layer.The experimental results show that the proposed model exhibits good fusion ability and achieves optimal performance, reaching 45.63%, 60.85%, 91.55%, and 0.564 in terms of Hit@1, Hit@10, Hit@30, and MeanRank, respectively, on the Zhu Nansun-Qian Boxuan dataset.Thus, an infertility knowledge graph reflecting the differences in syndrome differentiation and treatment is constructed to systematically establish the associations between the knowledge provided by multiple doctors.The findings of this work are highly relevant for the inheritance and development of personalized knowledge in Traditional Chinese Medicine(TCM).

Key words: knowledge graph, syndrome differentiation and treatment, infertility, probability soft logic, graph neural network

中图分类号:

TP391.1

马月坤, 张可心, 高唱. 体现辨证论治差异的不孕症知识图谱构建方法研究[J]. 计算机工程, 2023, 49(3): 280-287,295.

MA Yuekun, ZHANG Kexin, GAO Chang. Research on the Construction Method for Infertility Knowledge Graph Reflecting the Differences in Syndrome Differentiation and Treatment[J]. Computer Engineering, 2023, 49(3): 280-287,295.

https://www.ecice06.com/CN/Y2023/V49/I3/280

图/表 13

20230314190705

20230314190709

20230314190712

20230314190714

20230314190718

20230314190721

20230314190724

20230314190727

20230314190731

20230314190733

20230314190737

20230314190740

20230314190743

参考文献

[1] 方药中.辨证论治研究七讲[M].北京:人民卫生出版社, 1979. FANG Y Z.Seven lectures on syndrome differentiation and treatment[M].Beijing:People's Medical Publishing House, 1979.(in Chinese)
[2] 王映辉, 张润顺, 周雪忠, 等.名老中医经验共性规律及个性差异比较研究[J].世界科学技术(中医药现代化), 2009, 11(6):793-799. WANG Y H, ZHANG R S, ZHOU X Z, et al.Comparison of general rules and individual differences in clinical experience from the veteran TCM doctors[J].World Science and Technology (Modernization of Traditional Chinese Medicine and Materia Medica), 2009, 11(6):793-799.(in Chinese)
[3] 刘龙, 许玲, 李渡华, 等.应用模糊数学研究中医药的现状[J].中国临床医学, 2004, 11(6):934-936. LIU L, XU L, LI D H, et al.Study on the application of fuzzy mathematics method in traditional Chinese medicine research[J].Clinical Medical Journal of China, 2004, 11(6):934-936.(in Chinese)
[4] 秦亚莉, 郭少珍, 帅月圆.同病异证理论源流及其内涵[J].河南中医, 2019, 39(7):981-984. QIN Y L, GUO S Z, SHUAI Y Y.The origin and connotation of the theory of same disease and different syndromes[J].Henan Traditional Chinese Medicine, 2019, 39(7):981-984.(in Chinese)
[5] 邢梦, 邢玉瑞.中医病证关系研究评析[J].中华中医药杂志, 2018, 33(12):5290-5294. XING M, XING Y R.Research on the relationship between TCM diseases and syndromes[J].China Journal of Traditional Chinese Medicine and Pharmacy, 2018, 33(12):5290-5294.(in Chinese)
[6] 程指明."同病异治"与"同证异治"[J].黑龙江中医药, 1985, 14(2):12-13. CHENG Z M."Different treatment of the same disease" and "different treatment of the same syndrome"[J].Heilongjiang Journal of Traditonal Chinese Medicine, 1985, 14(2):12-13.(in Chinese)
[7] 刘凡.基于知识图谱技术的名老中医慢性胃炎辨证论治方案研究[D].北京:中国中医科学院, 2020. LIU F.Research on the veteran TCM physicians' scheme of syndrome differentiation and treatment of chronic gastritis based on knowledge graph[D].Beijing:China Academy of Chinese Medical Sciences, 2020.(in Chinese)
[8] 李新龙.中医师辨证论治失眠症知识图谱构建方法研究[D].北京:中国中医科学院, 2018. LI X L.Construction of knowledge graph of physicians' treatment based on syndrome differentiation for patients with insomnia[D].Beijing:China Academy of Chinese Medical Sciences, 2018.(in Chinese)
[9] 张雨琪, 李宗友, 王映辉, 等.赵炳南、朱仁康皮肤科流派用方经验知识图谱构建[J].中国中医药图书情报杂志, 2021, 45(2):1-5. ZHANG Y Q, LI Z Y, WANG Y H, et al.Construction of knowledge map on experience in TCM prescriptions of dermatology schools of Zhao Bingnan and Zhu Renkang[J].Chinese Journal of Library and Information Science for Traditional Chinese Medicine, 2021, 45(2):1-5.(in Chinese)
[10] 肖仰华.知识图谱:概念与技术[M].北京:电子工业出版社, 2019. XIAO Y H.Knowledge graphs:concepts and technologies[M].Beijing:Publishing House of Electronics Industry, 2019.(in Chinese)
[11] 王昊奋, 漆桂林, 陈华钧.知识图谱:方法、实践与应用[M].北京:电子工业出版社, 2019. WANG H F, QI G L, CHEN H J.Knowledge graph:methods, practices and applications[M].Beijing:Publishing House of Electronics Industry, 2019.(in Chinese)
[12] 张德政, 谢永红, 李曼, 等.基于本体的中医知识图谱构建[J].情报工程, 2017, 3(1):35-42. ZHANG D Z, XIE Y H, LI M, et al.Construction of knowledge graph of traditional Chinese medicine based on the ontology[J].Technology Intelligence Engineering, 2017, 3(1):35-42.(in Chinese)
[13] LI L, WANG P, YAN J, et al.Real-world data medical knowledge graph:construction and applications[J].Artificial Intelligence in Medicine, 2020, 103:101817.
[14] KIMMIG A, BACH S, BROECHELER M, et al.A short introduction to probabilistic soft logic[C]//Proceedings of the NIPS Workshop on Probabilistic Programming:Foundations and Applications.Cambridge, USA:MIT Press, 2012:1-4.
[15] PUJARA J, MIAO H, GETOOR L, et al.Using semantics and statistics to turn data into knowledge[J].AI Magazine, 2015, 36(1):65-74.
[16] WU Z H, PAN S R, CHEN F W, et al.A comprehensive survey on graph neural networks[J].IEEE Transactions on Neural Networks and Learning Systems, 2021, 32(1):4-24.
[17] SUN Z Q, HU W, ZHANG Q H, et al.Bootstrapping entity alignment with knowledge graph embedding[C]//Proceedings of the 27th International Joint Conference on Artificial Intelligence.Washington D.C., USA:IEEE Press, 2018:4396-4402.
[18] WANG Z C, LV Q S, LAN X H, et al.Cross-lingual knowledge graph alignment via graph convolutional networks[C]//Proceedings of 2018 Conference on Empirical Methods in Natural Language Processing.[S.l.]:ACL, 2018:349-357.
[19] 华一雄, 张执南.基于文本相似度和入出比的改进PageRank科研文献搜索方法[J].机械设计与研究, 2021, 37(1):6-9. HUA Y X, ZHANG Z N.Improved PageRank search method based on text similarity and in-out ratio[J].Machine Design & Research, 2021, 37(1):6-9.(in Chinese)
[20] 宫云宝, 甘亮, 黄九鸣.基于概率软逻辑模型的实体解析[J].计算机工程, 2017, 43(8):188-192, 199. GONG Y B, GAN L, HUANG J M.Entity resolution based on probabilistic soft logic model[J].Computer Engineering, 2017, 43(8):188-192, 199.(in Chinese)
[21] 陈泽东, 赵旭剑, 张晖, 等.面向开放式信息抽取系统的知识推理验证[J].西南科技大学学报, 2019, 34(4):72-80. CHEN Z D, ZHAO X J, ZHANG H, et al.Knowledge verification for open information extraction system[J].Journal of Southwest University of Science and Technology, 2019, 34(4):72-80.(in Chinese)
[22] XIE Y.MGAT:multi-view graph attention networks[J].Neural Networks, 2020, 132:180-189.
[23] VASWANI A, SHAZEER N, PARMAR N, et al.Attention is all you need[C]//Proceedings of the 30th Annual Conference on Neural Information Processing Systems.Cambridge, USA:MIT Press, 2017:5998-6008.
[24] 赵凯, 王华星, 施娜, 等.基于Neo4j桂枝汤类方知识图谱的研究与实现[J].世界中医药, 2019, 14(10):2636-2639, 2646. ZHAO K, WANG H X, SHI N, et al.Study and implementation on knowledge graph of Guizhi decoction associated formulas based on Neo4j[J].World Chinese Medicine, 2019, 14(10):2636-2639, 2646.(in Chinese)
[25] 陈善达, 夏帅帅, 邓文祥, 等.基于Neo4j的冠心病中医辨证论治知识图谱研究[J].中国医药导报, 2021, 18(21):138-141. CHEN S D, XIA S S, DENG W X, et al.Study on the knowledge map of Chinese medicine syndrome differentiation and treatment of coronary heart disease based on Neo4j[J].China Medical Herald, 2021, 18(21):138-141.(in Chinese)
[26] 李经纬.中医大辞典[M].北京:人民卫生出版社, 2004. LI J W.Dictionary of traditional Chinese medicine[M].Beijing:People's Medical Publishing House, 2004.(in Chinese)

选择文件类型/文献管理软件名称

选择包含的内容