ROUGH SET APPROACH TO INCOMPLETE INFORMATION SYSTEMS

摘要：

The rough set theory, based on the notion of the indiscernibility relation and defined by an equivalence relation, encounters difficulty in analyzing incomplete information systems where some values of attributes are unknown. In this paper we present two generalizations of the rough sets theory to distinguish two different semantics for incomplete information: the missing value semantics and the absent value semantics. The first generalization introduces the use of a compatibility relation in order to formalize the idea of absent value semantics. The second idea uses the valued similarity relations. A theoretical analysis and experiments show that for the valued similarity approach it is possible to obtain more informative approximations and decision rules. We compare with competing strategies theoretically and empirically and demonstrate the advantages of our proposed approaches.

关键词： Rough set Incomplete information Tolerance relation Compatibility relation Valued similarity relation Decision rules

作者: HONG TIAN XUDONG SONG

作者单位: Software Institute, Dalian Jiaotong University, Dalian, China

会议类型: 国际会议

会议名称: The Second International Conference on Information & Systems Sciences(ICISS2008)(第二届信息与系统科学国际会议)

会议地点: 大连

会议语种:英文

页码: 1011-1024

在线出版日期: 2008-12-18（万方平台首次上网日期，不代表论文的发表时间）