基于贝叶斯分类算法的治疗中风中药组方研究

Analyzing the Data of Chinese Medicine Used in Disease Treatment of Stroke Based on the Bayes Analysis

目的运用现代数据挖掘技术结合疗效数据对中医中风临床中使用的方剂进行分析,得到有价值的信息以支持中药新药的研发,并且获得一些有指导临床价值的数据分析结果。方法首先从中风诊疗文献库中抽取大量方剂和疗效信息,接着进行数据清洗,术语规范化等处理,然后运用数据挖掘中的贝叶斯等计算方法,进行关于治疗中风方药的知识发现,建立中药组方疗效模型,通过方剂、证候、药味的出现次数、有效率等关键数据进行统计计算。结果具有较好疗效的中药及其和证候、疗效之间的联系和规律,如赭石、胆星、赤芍、川芎在肝阳上亢证候使用时疗效好。结论这些结果可以为新药的研发提供新的技术支持,为临床医师用药提供参考和支持。

Objective Data mining technology was used in analyzing the data of Chinese medicine for disease treatment of stroke, some valuable information and knowledge areabtained for developing new medicine and doctor’decisionmaking.Methods Some treatment data and curative effect datafrom treatment of stroke were selected.Data was cleaned for further data analysis,like term standardization,datatransformation,ect.The Bayesian analysis technology was applied for knowledge discovery.Results The curative effect prediction model was buided based on the syndrome,medicine frequency,curative effect,the relationship between them and some valuable rules were finded in the research.For example in this paper Ocher,Danxing,Lactiflora,Chuanxiong Mulberry had high curative effect in treatment of hyperactivity of liver yang syndrome.Conclusion These result can support the new medicine discovery and help the doctor make decision.