Computer Engineering and Applications ›› 2008, Vol. 44 ›› Issue (14): 149-151.

• 数据库、信号与信息处理 • Previous Articles     Next Articles

PSOIDTAR method based on particle swarm optimization for attribution under incomplete decision-making table

ZENG Zheng-liang1,LUO Ke1,WANG Ying2   

  1. 1.Institute of Computer and Communication Engineering,Changsha University of Science and Technology,Changsha 410076,China
    2.Institute of Mathematics and Computer Science,Hunan Normal University,Changsha 410081,China
  • Received:2007-11-08 Revised:2008-02-27 Online:2008-05-11 Published:2008-05-11
  • Contact: ZENG Zheng-liang

基于粒子群的不完备决策表属性约简PSOIDTAR法

曾正良1,罗 可1,王 莹2   

  1. 1.长沙理工大学 计算机与通信工程学院,长沙 410076
    2.湖南师范大学 数学与计算机科学学院,长沙 410081
  • 通讯作者: 曾正良

Abstract: Attribution reduction is the core of the rough set theory.Because the theory of the classical rough set can’t apply to the incomplete decision-making table,the paper proposes a rough set attribute reduction algorithm based on the binary Particle Swarm Optimization(PSO),taking the attribution reduction to a part of the 0-1 combination optimization problem.The paper imports approximate classified precision and the approximate classified quality,and makes an efficient particle adaptive function in order to get a best reduction.Finally,experimental results show the algorithm can find the minimal relative reduction,and it is fast and effective.

Key words: rough set, incomplete decisionmaking table, Particle Swarm Optimization(PSO), attribute reduction, PSOIDTAR

摘要: 属性约简是粗糙集理论的一个核心部分。由于经典的粗糙集模型对不完备信息系统不适应,通过把属性约简问题归结为0-1组合优化问题,提出了一种应用二进制粒子群算法来求解属性约简的方法。通过引入近似分类精度和近似分类质量,为获得最小约简确定了有效合理的粒子适应度函数。仿真实验结果表明该算法能得到最小相对约简,且具有较高的运算效率。

关键词: 粗糙集, 不完备决策表, 粒子群算法, 属性约简, PSOIDTAR