邻域系统粗糙集和覆盖粗糙集  

Neighborhood System Based Rough Set and Covering Based Rough Set

作　　者：王丽娟[1,2] 吴陈[2] 杨习贝[1,2] 杨静宇[1] 

机构地区：[1]南京理工大学计算机科学与技术学院,南京210094 [2]江苏科技大学计算机科学与工程学院,镇江212003

出　　处：《计算机科学》2013年第1期221-224,共4页Computer Science

基　　金：国家自然科学基金(61100116);中国博士后科学基金(20100481149);江苏省自然科学基金(BK2011492);江苏省高校自然科学基金(11KJB520004);江苏省博士后科学基金(1101137C)资助

摘　　要：邻域系统粗糙集和覆盖粗糙集是经典粗糙集的两种重要扩展。通过分别比较各模型中下(上)近似集之间的包含关系和近似精度之间的大小关系,深入探讨邻域系统粗糙集和6种覆盖粗糙集模型之间的关系,得出了邻域系统粗糙集和6种覆盖粗糙集模型的下(或上)近似集之间的关系是明确的,其要么是可以比较的,要么是不可以比较的,证明了可比较的具有包含甚至等价关系,不可比较的通过反例进行了佐证。对不同扩展粗糙集的对比研究加深了对这些模型的理解,同时也为宏观上学习和认识粗糙集提供了帮助。Neighborhood system based rough set and covering based rough set are two important expansions of the classical rough set.By means of comparing the lower approximation sets,the upper approximation sets and accuracy measures,the relationships of neighborhood system based rough set and six covering based rough set models were systematically studied.The conclusion is that the relationships between the lower or upper approximations of neighborhood system based rough set and six covering based rough set models are clear,either can be compared,or not.Under the comparable situation,it was proved that there are inclusion even equivalence relations.Under the incomparable situation,it was proved by counter-examples.The comparative study on different expansions of rough set not only provides a better understanding of these models,but also gives a hand on learning rough set in the macroscopic level.

关 键 词：邻域系统粗糙集 覆盖粗糙集 近似精度 比较 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]