基于模糊粗糙集和蜂群算法的属性约简

来源期刊:中南大学学报(自然科学版)2013年第1期

论文作者:王世强 张登福 毕笃彦 张立东 王占领 李洋 雍霄驹

文章页码:172 - 178

关键词:属性约简;粗糙集理论;模糊粗糙集;依赖性;人工蜂群算法

Key words:attribute reduction; rough set theory; fuzzy rough sets; dependency; artificial bee colony algorithm

摘    要:为获取连续属性数据集的最小属性子集,提出一种基于模糊粗糙集和人工蜂群算法的约简方法。首先由边缘蕴含算子和t-模给出集合的模糊粗糙近似,以下近似构建模糊粗糙正域,并据此确定决策属性对条件属性集的依赖度,然后通过依赖度和约简率构建能够反映属性集大小和重要性的目标函数,将属性约简问题转化为优化问题,最后以目标函数为迭代准则,利用人工蜂群优化算法完成数据集的属性约简。仿真结果表明:该方法在不降低分类正确率的同时,可以有效降低属性维数。

Abstract: To acquire the minimum attribute reduction of the dataset with continual attribute values, a novel method was proposed on the basis of fuzzy rough sets (FRS) and artificial bee colony (ABC) algorithm. Firstly, the fuzzy rough approximation of the sets was given via a border implicator and a t-norm, and the fuzzy rough positive region and dependency of the decision attribute about the condition attribute have generated using this approximation. Then, an objective function indicating the importance and size of attribute sets was constructed based on the concept of dependency and reduction rate. Via this operation, the problem of attribute reduction was converted to that of optimization. Finally, the attribute reduction for datasets was performed using the ABC algorithm under the guidance of the objective function value. Experimental results show that the strategy can reduce the attribute dimensions efficiently without sacrificing the classification accuracy.

相关论文

  • 暂无!

相关知识点

  • 暂无!

有色金属在线官网  |   会议  |   在线投稿  |   购买纸书  |   科技图书馆

中南大学出版社 技术支持 版权声明   电话:0731-88830515 88830516   传真:0731-88710482   Email:administrator@cnnmol.com

互联网出版许可证:(署)网出证(京)字第342号   京ICP备17050991号-6      京公网安备11010802042557号