Abstract: Region’s differences of TCM’s culture lead to much uncertainty in TCM, to solve the data problem of decision support system for new medicine development based on data mining, a series of processing methods to standardize the original TCM data are proposed. Data reduction technology, clustering analysis and fuzzy set theory are applied to improve the quality of TCM data, getting important rules from the preprocessed TCM database, and providing powerful decision support for exploring new medicine.

Key words: Data preprocessing, Data mining, Data reduction, Fuzzy set, Membership function

摘要： 中药文化的地区差异带来了中医药数据的众多不确定性，为解决基于数据挖掘的新药研制决策支持系统的数据问题，提出了一套规范原始中医药数据的处理方法。应用了数据归约技术、聚类的方法、模糊集理论改进了中医药数据的质量，使得在预处理后的中药方剂数据库中成功挖掘出重要规则，为研制中药新药提供了有力的决策支持。

关键词: 数据预处理, 数据挖掘, 数据归约, 模糊集, 隶属函数

CLC Number: 

• TP274+.2