计算机科学 ›› 2014, Vol. 41 ›› Issue (7): 236-241.doi: 10.11896/j.issn.1002-137X.2014.07.049

• 软件与数据库技术 • 上一篇    下一篇

代数约简的条件信息熵表示及其高效约简算法

黄国顺,曾凡智,文翰   

  1. 佛山科学技术学院理学院 佛山528000;佛山科学技术学院电子信息工程学院 佛山528000;佛山科学技术学院理学院 佛山528000
  • 出版日期:2018-11-14 发布日期:2018-11-14
  • 基金资助:
    本文受广东省自然科学基金资助

Conditional Information Entropy Representation of Algebraic Reduction and its Efficient Algorithm

HUANG Guo-shun,ZENG Fan-zhi and WEN Han   

  • Online:2018-11-14 Published:2018-11-14

摘要: 给出如何保持正区域不变的语义分析,提出一种修正条件信息熵计算公式,证明保持修正条件信息熵不变与保持正区域不变相互等价。在此基础上,给出代数约简概念的修正条件信息熵表示。给出反例说明修正条件信息熵不具有单调性,导致没法给出自底向上的启发式约简算法,证明了代数协调集中不可删除属性的不可逆性质,提出一种自顶向下直接删除属性的高效约简算法。它从所有条件属性集出发,逐步删除不必要的属性,只需遍历各属性一次,即可保证得到原始决策表的一个代数约简。数值算例和实验验证了该算法的正确性和高效性。

关键词: 条件信息熵,正区域,代数约简,算法 中图法分类号TP18文献标识码A

Abstract: A semantic analysis on how to keep positive region unchanged was given.An improved conditional information entropy was proposed.It was proved that remaining the modified conditional information entropy unchanged and remaining positive region unchanged are equivalent.Therefore,some main concepts of algebraic reduction were described by the revised conditional information entropy.However,a counter example illustrates that its monotonicity does not hold,which means a heuristic reduction algorithm can not be constructed based on bottom-up.Any attribute in an algebraic consistent set is not irreversible if it is checked unsuppressible.An efficient algorithm based on top-down was proposed,which starts from condition attribute set,removes the unnecessary attribute step by step.It is finally guaranteed to obtain an algebraic reduction by traversing the attributes only once.Numerical example and the experimental results show that the algorithm is valid and efficient.

Key words: Conditional information entropy,Positive region,Algebraic reduction,Algorithm

[1] Pawlak Z.Rough set [J].Commucation of the ACM,1995,38(11):89-95
[2] Hu X H,Cerene N.Learning in relational databases:a rough set approach [J].International Journal of Computation Intelligence,1995,11(2):323-338
[3] 王国胤,于洪,杨大春.基于条件信息熵的决策表约简[J].计算机学报,2002,25(7):759-766
[4] Wang G Y,Zhao J,An J J,et al.A comparative study of algebra viewpoint and information viewpoint in attribute reduction [J].Fundamenta Informaticae,2005,68(3):289-301
[5] Liang J Y,Wang F,Dang C Y,et al.An efficient rough feature selection algorithm with a multi-granulation view [J].International Journal of Approximate Reasoning,2012,53(6):912-926
[6] 陈玉明,吴克寿,谢荣生.基于相对知识粒度的决策表约简[J].山东大学学报:工学版,2012,2(6):8-12
[7] Qian Y H,Liang J Y,Pedrycz W,et al.Positive approximation:an accelerator for attribute reduction in rough set [J].Artificial Intelligence,2010,174(9/10):597-618
[8] 葛浩,李龙澍,杨传健.新的可分辨矩阵及其约简方法[J].控制与决策,2010,5(12):1891-1895
[9] 刘启和,李凡,闵帆,等.一种基于新的条件信息熵的高效知识约简算法[J].控制与决策,2005,0(8):878-882
[10] 马胜蓝,叶东毅.信息熵最小约简问题的若干随机优化算法[J].模式识别与人工智能,2012,25(1):96-104
[11] 钱文彬,杨柄儒,徐章艳,等.基于信息熵的核属性增量式高效更新算法[J].模式识别与人工智能,2013,26(1):42-49
[12] 蒋云良,杨章显,刘勇.不协调信息系统快速属性分布约简方法[J].自动化学报,2012,38(3):382-388
[13] 钱进,叶飞跃,孟祥萍,等.一种基于新的条件信息量的属性约简算法[J].系统工程与电子技术,2007,9(12):2154-2157
[14] 黄国顺,刘云生.不一致决策表信息熵约简与代数约简的核计算与转化[J].小型微型计算机系统,2008,29(2):308-312
[15] 黄国顺,刘云生.不一致决策表各种属性约简的不一致性分析与转化[J].小型微型计算机系统,2008,29(4)703-708
[16] 徐章艳,刘作鹏,杨炳儒,等.一个复杂度为max(O(|C||U|),O(|C|2|U/C|))的快速属性约简算法[J].计算机学报,2006,29(3):391-399
[17] Blake C L,Merz C J.UCI Repository of Machine Learning Databases,University of Califonia at Irvine.http://www.ics.uci.edu/~mlearn/MLRepositioy.html,1998
[18] Liang J Y,Qian Y H.Information granularity and entropy theoryin information system [J].Science in China,(F),2008,51(10):1427-1444

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!