基于知识粒度的决策空间方法研究与实现

本文利用粗糙集理论中的粗度概念，在对获取规则进行扩展性的粒度描述的基础上，提出了一种工具——知识依赖关系图（KDRD），构造了n维知识粒度决策空间以及基于多决策类的同构映射决策空间，建立了基于KDRD和决策空间的决策算法，并结合机械故障诊断说明了本方法非常适于故障诊断等通过结论获取原因的智能决策系统．

Research on Decision Space Modeling and Implementation Based on Knowledge Granularity

In this paper, granularity in rough set theory is used to describe rules based on the concept of rough degree. A tool for knowledge dependent relation diagram (KDRD) is put forward. Then dimensional knowledge granularity decision space and the same construction mapping decision space based on multi-decision sorts is established, and decision algorithm based on KDRD and decision space is introduced. Finally, through combining with the technique of fault diagnosis ,it is proved that the approach is very appropriate to the intelligent decision system.