计算机工程与应用 ›› 2014, Vol. 50 ›› Issue (2): 107-111.

• 数据库、数据挖掘、机器学习 • 上一篇    下一篇

随机覆盖目标信息系统的属性约简

方连花,李克典   

  1. 漳州师范学院 数学与信息科学系,福建 漳州 363000
  • 出版日期:2014-01-15 发布日期:2014-01-26

Attribute reduction in random covering objective information system

FANG Lianhua, LI Kedian   

  1. Department of Mathematic and Information Science, Zhangzhou Normal University, Zhangzhou, Fujian 363000, China
  • Online:2014-01-15 Published:2014-01-26

摘要: 引入随机覆盖目标信息系统的概念,以证据理论中的信任测度和似然测度为基本工具,研究了协调随机覆盖目标信息系统的属性约简和不协调随机覆盖目标信息系统的正域约简问题,最后给出实例验证了约简方法的有效性。

关键词: 随机覆盖信息系统, 信任测度, 似然测度, 正域约简, 辨识矩阵

Abstract: The concept of random covering objective information system is given The attribute reduction is studied in consistent random covering information system, and the positive domain reduction is discussed in inconsistent random covering information system, by means of the belief measure and plausibility measure in the evidence theory. An example is used to illustrate the validity of this method.

Key words: random covering information system, belief measure, plausibility measure, positive domain reduction, discernable matrix