计算机工程与应用 ›› 2009, Vol. 45 ›› Issue (25): 132-134.DOI: 10.3778/j.issn.1002-8331.2009.25.040

• 数据库、信息处理 • 上一篇    下一篇

粒计算中决策规则的提取

徐久成1,史进玲1,2,成万里1   

  1. 1.河南师范大学 计算机与信息技术学院 智能信息处理重点实验室,河南 新乡 453007
    2.许昌学院 国际教育学院,河南 许昌 461000
  • 收稿日期:2008-10-22 修回日期:2008-12-26 出版日期:2009-09-01 发布日期:2009-09-01
  • 通讯作者: 徐久成

Algorithm for decision rules extraction based on granular computing

XU Jiu-cheng1,SHI Jin-ling1,2,CHENG Wan-li1   

  1. 1.Key Lab for Intelligent Information Processing,College of Computer and Information Technology,Henan Normal University,Xinxiang,Xuchang,Henan 453007,China
    2.International School of Education,Xuchang University,Xuchang,Henan 461000,China
  • Received:2008-10-22 Revised:2008-12-26 Online:2009-09-01 Published:2009-09-01
  • Contact: XU Jiu-cheng

摘要: 为从决策表中获取最简决策规则,在研究粒计算理论的基础上,提出了一种基于粒计算的决策规则提取算法。该算法通过引入λ阶粒库的概念,利用粒计算的方法对由同一决策表所生成的粒库,从不同的角度和多个层次进行研究,尝试从较低阶的粒库中提取出尽可能多的具有一定规则覆盖度与置信度的简洁规则。最后通过实例验证了该算法的有效性。

关键词: 粒语言, 粒计算, 粒库, 决策规则

Abstract: To extract compact decision rules from decision table,an algorithm for decision rules extraction based on granular computing is proposed.In the algorithm,a concept of λ-rank granular base is defined,and granular bases of different ranks for the same decision table are brought out,using the methods of granular computing as many as possible,to extract decision rules which satisfy certain coverage and confidence limit of rules.Lastly,the validity for the algorithm is proved by analyzing examples.

Key words: granular language, granular computing, granular base, rules extraction

中图分类号: