嵌套决策粒度约简关系的研究  

Research on Reduction Relationships of Nested Granularities

作　　者：李道国[1] 苗夺谦[1] 张年琴[1] 

机构地区：[1]同济大学计算机科学与技术系

出　　处：《计算机科学》2006年第2期178-181,共4页Computer Science

基　　金：国家自然科学基金项目(60175016;60475019)

摘　　要：在基于粗糙集理论的知识发现中,知识约简是核心内容。因此,本文从理论上分析了相容决策表中嵌套决策粒度约简的关系,得出并证明了粗决策粒度的相对 D 核一定是细决策粒度相对 D 核的子集,粗决策粒度的一个相对D约简在满足相容性的条件下一定可以扩张成细决策粒度的一个约简。反之,细决策粒度的一个约简一定可以缩减为粗决策粒度的一个约简。研究结果对基于决策表的知识约简和知识发现有一定的实际意义。Knowledge Reduction plays an important role in knowledge discovery based on rough set theory. In this paper, therefore, the reduction relationships of nested granularities are theoretically analyzed in consistent information tables. We educe and prove that relative core of coarser decision granularity must be a subset of relative core of finer deciT sion granularity. At the same time,a relative reduction of coarser decision granularity that satisfies consistency can be extended into a relative reduction of its finer decision granularity. Inversely, a relative reduction of finer decision granularity can also be cut back a relative reduction of its coarser decision granularity. The conclusions are helpful for knowledge reduction and knowledge discovery based on rough sets.

关 键 词：相对D核 相对D约简 包客强度 偏序关系 粒划分 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TP391.41[自动化与计算机技术—控制科学与工程]