北京邮电大学学报

  • EI核心期刊

北京邮电大学学报 ›› 2011, Vol. 34 ›› Issue (6): 55-58.doi: 10.13190/jbupt.201106.55.wangh

• 论文 • 上一篇    下一篇

基于等价划分与蚁群优化的并行属性约简算法

王慧1,2,王京3,4   

  1. 1. 北京科技大学
    2. 中国人民公安大学
    3.
    4. 北京科技大学自动化学院
  • 收稿日期:2011-06-02 修回日期:2011-07-15 出版日期:2011-12-28 发布日期:2011-10-18
  • 通讯作者: 王慧 E-mail:wanghui_wwh@163.com
  • 作者简介:王慧(1973-),女,博士生,E-mail:wanghui_wwh@163.com 王京(1948-),男,教授,博士生导师
  • 基金资助:
    国家高技术研究发展计划项目

An Attribute Reduction Algorithm Based on Partition and Ant Colony Optimization

  • Received:2011-06-02 Revised:2011-07-15 Online:2011-12-28 Published:2011-10-18
  • Contact: WANG Hui E-mail:wanghui_wwh@163.com

摘要:

为降低经典信息熵属性约简算法的时间复杂度,在论证信息熵属性约简与论域对象划分细化约简等价的基础上,提出将蚁群并行优化处理机制引入划分细化约简过程的思想,蚁群搜索过程将属性重要性度量融入状态转移及信息素更新策略以对每次约简结果进行优化。通过复杂性分析与实例验证,该算法更适于大容量数据表的属性约简,可有效避免蚁群搜索的盲目性并在较小迭代规模下快速获得约简集。

Abstract:

To decrease the time complexity of the attribute reduction algorithm based on information entropy, the concurrent processing of ants colony optimization is introduced into partition. This process based on the judgement of attribute reduction between partition and information entropy. In this searching process the strategy of state transfer and pheromone update reflect the difference of the attribute’s importance in order to optimize the results. By the analysis of this algorithm’s complexity and the example’s application,it is shown that the algorithm can avoid the blindness of the ant colony searching and reduce the size of the iteration to obtain reduction faster.This is more suitable for large data base.

中图分类号: