摘要
目的:本研究基于二分图对疾病和中药之间的关联性进行分析,构成"疾病?中药"的复杂网络模型。方法:首先,根据每种疾病对应方剂中的中药分别构建疾病数据库与中药数据库,计算出每种疾病和中药所对应的度数;其次,运用Jaccard相似度函数建立中药关联性,并根据中药关联性推断出疾病关联性;最后,利用二分图和复杂网络方法对疾病与中药之间的关系进行分析,并用PAJEK软件进行仿真实验。结果:得出疾病与中药、疾病与疾病、中药与中药、证候与中药之间的二分图关系。结论:通过对实验结果进行分析,得出药物和疾病相关性结论,并进一步提出在以后研究中采用二叉树建立"疾病?中药"数据库,以便后期对数据进行关联分析。
Bipartite graph is a special model in the graph theory with the characteristics of clustering and associativity. In this study, the analysis of the correlation between diseases and herbal medications was performed based on bipartite graph before constructing the ＂Disease-Herb＂ complex network model. Firstly, the disease database and herb database were established in which the diseases and their prescriptions were involved. After that, the degrees of diseases and herbs were calculated separately. Secondly, the correlation among herbs was analyzed through the ＂Jaccard＂ similarity function, and so was the correlation of diseases. Finally, the correlation between diseases and herbal medications was deduced via the methods of bipartite graph and complex network; and the simulate experiment was implemented by ＂Pajek＂ software. As a result, the correlations of diseases and herbal medications, diseases and diseases, herbs and herbs, and syndromes and herbs were obtained. It was concluded that the correlation between diseases and herbal medications can be drove through data analysis. The binary tree method should be reasonably used to establish a ＂Disease-Herb＂ database in the future to analyze the correlations conveniently.
出处
《世界科学技术-中医药现代化》
2016年第4期575-581,共7页
Modernization of Traditional Chinese Medicine and Materia Medica-World Science and Technology
基金
2014年湖北中医药大学校级教学研究项目(2014B17):医学信息工程专业学生数据建模能力培养模式研究
负责人:胡芳
2014湖北省教育厅科学研究计划项目(D20152003):基于肝病的中医临床术语本体构建研究
负责人:马利
关键词
疾病
方剂
中药
二分图
数据建模
Disease
prescription
herb
bipartite graph
data modeling
