基于相似度的变精度容差关系扩充粗糙集模型

Extension of rough set model based on similarity of variable precision dynamic tolerance relation

比较分析了已有扩充粗糙集模型,提出基于相似度的变精度容差关系扩充粗糙集模型。该模型根据两对象的相似度及属性值缺失情况划分容差类。通过与基于容差关系、相似关系、限制容差关系等扩充粗糙集模型比较,证实了该模型对不完备信息系统处理的现实可行性和有效性,最后给出了相对正域的概念及属性约简规则。

An extended rough set model based on similarity of variable precision dynanxic tolerance relation is proposed on the basis of comparing the existing extended rough set models. This model divides the tolerance classes according to the similarity of two objects and missing property value. The realistic feasibility and effectiveness of the model to handle the incomplete information systems is con- firmed through the comparison among the expanded rough set models of tolerance relation, similarity relation, limited tolerance rela- tion. At the end, the concept of relative positive region and attribute reduction rule are given.