Computer Engineering and Applications ›› 2007, Vol. 43 ›› Issue (14): 38-41.

• 学术探讨 • Previous Articles     Next Articles

Knowledge granularity and the relationship between knowledge granularities in incomplete information systems

  

  • Received:2006-06-05 Revised:1900-01-01 Online:2007-05-10 Published:2007-05-10

基于粗糙集的知识粒度及粒度关系研究

王树锋 吴耿锋 潘建国   

  1. 上海大学计算机学院 上海大学计算机学院
  • 通讯作者: 王树锋

Abstract: In this paper, we propose a definition of knowledge granular in incomplete information system based on rough set, and study the relationship of knowledge granularity and information entropy. We also propose the mutual support degree of knowledge and mutual different degree of knowledge and study the relation between them in order to provide a viable method of measurement of knowledge and evaluation of knowledge. Understanding the measurement of knowledge and evaluation of the similar degree of knowledge, on the one hand, certain granular of knowledge based on knowledge of granularity can be extracted from an incomplete information system; on the other hand, by comparing the difference support degree of knowledge bases which have been integrated, we can indirectly assess the complete of knowledge.

Key words: granular computing, knowledge granularity, mutual support degree, difference support degree

摘要: 本文基于粗糙集理论,针对不完备信息系统,提出了一种知识粒度的定义,并研究了知识粒度与信息熵的关系;提出了知识的相互支持度和知识的差异支持度的概念,并研究了二者之间的关系,为进行知识的度量和知识的评价提供了一种可行的方法。弄清楚知识的度量和知识相似程度的评价,一方面可以根据知识的粒度,从不完备信息系统中抽取符合一定粒度的知识;另一方面当进行知识融合时,可以比较融合后形成的知识库之间的差异相似程度,间接评价知识的完备性。

关键词: 粒度计算, 知识粒度, 相互支持度, 差异支持度