摘要
在决策属性已知、条件属性值分布不确定的情况下,用基于概率相似度原理和按决策属性划分系统的原则,对缺损数据进行填补,可使不完备决策信息系统的完备化具有较高可信度。
Based on probability similarity and the system partition according to decision attribute in the condition of unknown data distribution,this paper put forward a new method to improve the algorithm of ROUSTIDA,which made a high reliability for the missing data of incomplete decision information system.
出处
《计算机应用研究》
CSCD
北大核心
2009年第3期881-883,共3页
Application Research of Computers
基金
河北省教育厅计划资助项目(2007205)
关键词
粗糙集
缺损数据
概率相似度
算法
rough set
missing data
probability similarity
algorithm
