基于模糊聚类与模糊关联的中药处方剂量配伍规律研究

Research on Association Rule Mining Method of Prescription Dosage Based on Fuzzy Clustering and Fuzzy Association

目的:探究中药处方剂量大小分布形式,挖掘中药处方剂量关联关系,为临床用药提供参考,为中药处方量效研究提供方法基础。方法:在对12 138组临床中药处方进行预处理后,使用模糊聚类与模糊关联规则进行挖掘。结果:研究求得309味药物大中小剂量中心点及其各自剂量隶属度矩阵,据此生成21条剂量模糊关联规则,并将该算法与传统数据挖掘方法进行比较。结论:模糊聚类算法可实现中药剂量的合理划分,模糊关联算法较之传统关联分析手段更适用中药处方剂量的挖掘。

Objectives:To explore dosage size distribution of Chinese medicine prescription and excavate the association relationship of prescription dosage so as to provide reference for clinical medication and method foundation for researching prescription dosage as well as effect.Methods:Based on the pretreatment of 12138 groups of clinical Chinese medicine prescriptions,fuzzy clustering and fuzzy association rules were used for mining.Results:The central points of309 kinds of medicine dosage sizes along with their respective dosage membership matrixes were acquired,which were applied to generate 21 dosage fuzzy association rules.Meanwhile,this algorithm was used to compare with traditional data mining method.Conclusions:Fuzzy clustering algorithm can realize the reasonable partition of Chinese medicine dosage,and fuzzy association algorithm,compared with traditional association analysis means,is more suitable for excavating prescription dosage.