论文部分内容阅读
属性约简是粗糙集理论的重要研究内容,本文基于模糊信息系统,一方面,通过模糊相似关系定义了条件相似度以及决策相似度,建立了关于条件相似度与决策相似度的相对比较矩阵,给出了属性约简集的新定义;另一方面,结合知识的粒度、分辨度、关联度确定了条件属性对决策属性的重要度,由此,提出了一种基于相似度比较的模糊属性约简方法。
Attribute reduction is an important research content of rough set theory. Based on fuzzy information system, on the one hand, conditional similarity and decision similarity are defined by fuzzy similarities, and a relative comparison matrix about the similarity of conditional similarity and decision-making is established. A new definition of attribute reduction set is given. On the other hand, the granularity, resolution and relevance of knowledge determine the importance of conditional attributes to decision attributes. Thus, a fuzzy attribute based on similarity comparison Reduction method.