摘要
针对目前数据挖掘隐私保护中信息系统属性分类问题,基于粗糙集理论提出了一种信息系统中属性划分方法.首先对信息系统中的易获知属性和敏感属性进行定义和描述,然后在属性依赖度和属性重要度的基础上,提出了信息系统中属性的敏感系数的定义.实验结果表明:该方法简单、直观,对信息系统进行属性划分与属性约简得到的结果是一致的.
In the light of Classification of Attribute in Information System in the data mining of privacy preserving,a method of attribute classification had been proposed,based on rough set theory.First,easily-knowing attribute and sensitive attribute had been defined and described in the method.Second,based on the attribute dependence and attribute importance,the definition of attribute sensitivity coefficient in the information system had been proposed.The experimental results show that the method is simple,intuitive and the attribute classification with the method in the information system is consistent with the result of the attribute reduction.
出处
《中北大学学报（自然科学版）》
CAS
北大核心
2012年第6期695-698,共4页
Journal of North University of China(Natural Science Edition)
关键词
粗糙集
属性重要度
依赖性
属性划分
rough set theory
significance
dependence
attribute classification
