Abstract: A definition of attribute significance is proposed based on knowledge rough entropy in ordered information systems, and important properties are obtained. It can be found that using the definition can measure uncertainty of an attribute set in the ordered information systems. A heuristic algorithm for attributes reduction is acquired in the systems. An example illustrates the validity of this algorithm, and results show that the algorithm is an efficient tool for data mining, and provides an important theoretical basis for knowledge discovery in ordered information systems.

Key words: rough set, information system, dominance relation, heuristic algorithm

摘要： 基于序信息系统的知识粗糙熵，在系统中引入属性重要性的概念，利用该测度能度量序信息系统中属性集的不确定性，基于此，提出序信息系统中基于知识粗糙熵的启发式约简算法。通过实例对该方法的有效性进行检验，结果显示该算法可以作为一种有效的数据挖掘工具，为序信息系统的知识发现提供理论基础。

关键词: 粗糙集, 信息系统, 优势关系, 启发式算法

CLC Number: 

• TP311