决策表属性约简的相对信息量表示  

Relative Information Quantity Representation for Attribute Reduction of Decision Tables

作　　者：胡逢彬[1] 桂现才[2] 

机构地区：[1]湛江师范学院商学院,广东湛江524048 [2]湛江师范学院数学与计算科学学院,广东湛江524048

出　　处：《计算机技术与发展》2006年第7期39-41,共3页Computer Technology and Development

基　　金：湛江师范学院科研基金(W0428)

摘　　要：知识约简是粗糙集理论研究的主要内容之一,该文在信息系统中引入了知识的相对信息量的概念。对一致决策表,证明了其属性约简在代数表示下和相对信息量表示下是等价的,但对不一致决策表,举例说明其属性约简的代数表示不能用相对信息量来等价表示。由此可见,相对信息量表示比代数表示直观,但不能完全代替代数表示方法。Reduction of knowledge is one of the important topics in the research on the rough set theory. In this paper, the concept of the relative information quantity of decision table is introduced in the information system. In a consistent decision table, the equivalence properties between algebraic representation and relative information quantity representation of attribute reduction are proved. Through examples, it shows that attribute reduction of an inconsistent decision table cannot entirely be represented by relative information quantity. This shows that relative information quantity representation is more visual than algebraic representation ,but it can＇t displace absolutely each other.

关 键 词：粗糙集 决策表 相对信息量 互信息 属性约简 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]