双精度容差关系决策粗糙集及其属性约简

摘要

Based on rough set theory and Bayes theory,it expended the decision-theoretic rough set.In order to research decision-theoretic rough set in incomplete interval-valued information system,first,it came up with the similarity degree and proposed the dual-variable precision tolerance relation.Then,based on the relation,it established the decision-theoretic rough set model in the incomplete interval-valued information system.Then it redefined the definition of uncertainty measure and studied the reduction of the model.Finally,it proposed the reduction algorithm and applied to a target recognition example.%决策粗糙集理论是经典粗糙集理论结合贝叶斯理论拓展而来.为在不完备区间值信息系统下研究决策粗糙集理论,首先提出属性相似度的概念,并基于属性相似度定义了双精度容差关系.然后,基于双精度容差关系,结合贝叶斯最小风险决策原则建立不完备区间值信息系统下决策粗糙集模型.针对该模型,对不确定性度量进行修正并基于修正的不确定性度量对该模型的属性约简进行研究.最后提出属性约简算法并应用于目标识别实例.