计算机工程与应用 ›› 2015, Vol. 51 ›› Issue (14): 95-98.

• 数据库、数据挖掘、机器学习 • 上一篇    下一篇

变精度粗集模型的属性约简研究

汪小燕,杨思春,申元霞   

  1. 安徽工业大学 计算机科学与技术学院,安徽 马鞍山 243032
  • 出版日期:2015-07-15 发布日期:2015-08-03

Research of attribute reduction in variable precision rough set model

WANG Xiaoyan, YANG Sichun, SHEN Yuanxia   

  1. School of Computer Science & Technology, Anhui University of Technology, Ma’anshan, Anhui 243032, China
  • Online:2015-07-15 Published:2015-08-03

摘要: 针对大量存在的不一致决策表,研究了分配量函数和[β]分配量函数定义。依据Ziarko变精度粗糙集模型,提出利用计算条件属性组合的[β]重要度来选择属性的[β]重要度属性约简和利用[β]二进制可辨矩阵实现的[β]分配量属性约简,可有效解决不一致决策表属性约简问题。

关键词: 粗糙集, 属性约简, 变精度, 二进制可辨矩阵

Abstract: The distribution quantity function and [β] distribution quantity function are researched aiming at many inconsistent decision tables. Based on Ziarko variable precision rough set model, it puts forward that attribute reduction about [β]importance utilizing to calculate the importance of condition attributes combination to select the attribute and attribute reduction about [β]distribution quantity by [β] binary discernable matrix. They can effectively solve the problem of attribute reduction in inconsistent decision tables.

Key words: rough set, attribution reduction, variable precision, binary discernable matrix