论文部分内容阅读
粗集理论是处理知识不精确和不完善的一种归纳学习方法,其基本思想是在保持分类能力不变的前提下,通过知识约简,导出概念的分类规则。熵作为对不确定性的一种度量,可用于描述近似空间(U,R)中对象的分类情况。在文中,知识的粗糙性定义为近似空间中的粗糙熵,近似空间上基于等价关系的划分过程是其粗糙熵不断减小的过程。同时讨论了信息系统中的若干粗糙熵性质。
Rough set theory is a kind of inductive learning method that deals with inaccurate and imperfect knowledge. Its basic idea is to deduce the classification rules of concepts through knowledge reduction while keeping the classification ability unchanged. Entropy as a measure of uncertainty can be used to describe the classification of objects in approximate space (U, R). In this paper, the roughness of knowledge is defined as the rough entropy in the approximate space. The partition process based on the equivalence relation in the approximate space is the process of decreasing its rough entropy. Several rough entropy properties in information systems are also discussed.