基于粗糙集与蚁群优化算法的特征选择方法研究倡.pdf

第 28 卷第 7 .7
2011 年 7 月 Vol .2011No
Application Research puters Jul
基于粗糙集与蚁群优化算法的
特征选择方法研究倡
吴克寿, 陈玉明, 谢荣生, 王晓栋
(厦门理工学院计算机科学与技术系, 福建厦门 361024)
摘要: 已有的基于蚁群优化算法的特征选择方法是从随机点出发,寻找最优的特征组合。讨论和分析了粗糙
集理论中的特征核思想,结合蚁群优化算法的全局寻优特点,以特征重要度作为启发式搜索信息,提出从特征核
出发基于粗糙集理论与蚁群优化的特征选择算法,简化蚁群完全图搜索的规模。在标准数据集上进行测
UCI
试,实验验证了新算法对于特征选择的有效性。
关键词: 粗糙集理论; 知识约简; 特征选择; 蚁群优化
中图分类号: 391 文献标志码: 文章编号: 1001唱3695(2011)07唱2436唱03
TP A
: /. .
doi j issn
Rough sets and ant colony optimization based feature selection
唱, 唱, 唱, 唱
WU Ke shou CHEN Yu ming XIE Rong sheng WANG Xiao dong
( Science Technology, Xiamen University of Technology, Xiamen Fujian 361024, China)
&
Abstract: 唱, 唱
. Many existing ACO based feature selection algorithms start from a random dot which aim at finding, the optimal fea
tures This thesis analyzed the feature core method of, rough sets and the global optimization ability of ACO proposed. a new
rough set approach to feature, selection based on ACO which adopted feature significance. as heuristic information The approach,
started from the feature core which changed the. complete graph to a smaller one To verify the efficiency of algorithm carried唱
out experiments on some standard UCI datasets. The results demonstrate that the proposed algorithm can provide efficient solu
tion to find a minimal s