计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (17): 136-138.DOI: 10.3778/j.issn.1002-8331.2010.17.038

• 数据库、信号与信息处理 • 上一篇    下一篇

基于模糊粗糙依赖度的连续值属性约简

翟兴隆,李续武   

  1. 空军工程大学 导弹学院 计算机工程系,陕西 三原 713800
  • 收稿日期:2009-12-24 修回日期:2010-03-29 出版日期:2010-06-11 发布日期:2010-06-11
  • 通讯作者: 翟兴隆

Continuous attribute reduction based on fuzzy rough dependence degree

ZHAI Xing-long,LI Xu-wu   

  1. Department of Computer Engineering,Missile Institute,Air Force Engineering University,Sanyuan,Shaanxi 713800,China
  • Received:2009-12-24 Revised:2010-03-29 Online:2010-06-11 Published:2010-06-11
  • Contact: ZHAI Xing-long

摘要: 针对传统的离散化技术所造成的信息丢失问题,提出了利用模糊粗糙集理论来进行属性约简的方法。描述了模糊等价关系下的粗糙集模型,定义了正域、依赖度等概念,提出了基于模糊粗糙依赖度的属性约简算法,该方法比传统属性约简方法具有更好的时间复杂性,并用实例证明了该算法的可行性。

关键词: 模糊粗糙集, 依赖度, 连续值, 属性约简

Abstract: As to the problem of information loss in the process of discretization,a method of attribute reduction based on the fuzzy rough sets theory is described.A model of rough sets under fuzzy equivalent relations is systematically investigated.The definitions of positive field,dependence degree are given.And an attribute reduction algorithm based on fuzzy rough dependence degree is particularly analyzed,which is more efficient than the traditional methods.Experiments show the feasibility of the application of the algorithm.

Key words: fuzzy rough sets, dependence degree, continuous features, attribute reduction

中图分类号: