作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程

• 人工智能及识别技术 • 上一篇    下一篇

基于二进制区分矩阵的增量式属性约简算法

丁棉卫 1,张腾飞 1,马福民 2   

  1. (1.南京邮电大学 自动化学院,南京 210023; 2.南京财经大学 信息工程学院,南京210023)
  • 收稿日期:2015-11-26 出版日期:2017-01-15 发布日期:2017-01-13
  • 作者简介:丁棉卫(1991—),男,硕士研究生,主研方向为粗糙集理论;张腾飞、马福民,副教授、博士。
  • 基金资助:
    国家自然科学基金(61105082,61403184);江苏省“青蓝工程”基金(QL2016);南京邮电大学“1311人才计划”项目(NY2013);南京邮电大学科研项目(NY215149)。

Incremental Attribute Reduction Algorithm Based on Binary Discernibility Matrix

DING Mianwei 1,ZHANG Tengfei 1,MA Fumin 2   

  1. (1.College of Automation,Nanjing University of Posts and Telecommunications,Nanjing 210023,China; 2.College of Information Engineering,Nanjing University of Finance and Economics,Nanjing 210023,China)
  • Received:2015-11-26 Online:2017-01-15 Published:2017-01-13

摘要: 增量式属性约简算法是动态数据挖掘技术的重要研究内容。为降低区分矩阵的存储空间,结合二进制区分矩阵便于计算以及形象直观的优点,给出一种压缩二进制区分矩阵的方法。将二进制区分矩阵的存储空间从|C|+1列简化成3列。通过动态更新二进制区分矩阵实现增量式求核,并以核为出发点,提出一种的增量式属性约简算法。通过实例计算及仿真实验验证了该算法的有效性。

关键词: 粗糙集, 增量式属性约简, 二进制区分矩阵, 核属性, 属性频率

Abstract: Incremental attribute reduction algorithm is one of the important research contents in the area of dyanmic data mining.To reduce the storage space of binary discernibility matrix,and by combining the advantages of the binary discernibility matrix that it facilitates the calculation and is visual,this paper proposes a method to compress binary matrix.It simplifies the binary discernibility matrix storage space from |C|+1 column to three columns.Through dynamically updating the binary discernibility matrix to incrementally get core.According to the core,the paper proposes an incremental attribute reduction algorithm based on binary discernibility matrix.Example calculation and experimental simulation prove the effectiveness of the algorithm.

Key words: rough sets, incremental attribute reduction, binary discernibility matrix, core attribute, attribute frequency

中图分类号: