###
工程科学与技术:2014,46(3):89-94
←前一篇   |   后一篇→
本文二维码信息
码上扫一扫!
水平划分决策表的属性约简算法
(1.安徽大学 计算智能与信号处理教育部重点实验室;2.滁州学院 机械与电子工程学院;3.安徽大学 计算机科学与技术学院;4.滁州学院 计算机与信息工程学院)
An Algorithm for Attribute Reduction Based on Horizontally Partitioning Decision Table
(1.Key Lab. of Computation Intelligence and Signal Processing of Education Ministry,Anhui Univ.;2.School of Mechanical and Electronic Eng.,Chuzhou Univ.;3.School of Computer Sci. and Technol.,Anhui Univ.;4.School of Computer and Info. Eng.,Chuzhou Univ.)
摘要
图/表
参考文献
相似文献
附件
本文已被:浏览 2702次   下载 88
投稿时间:2013-10-14    修订日期:2013-12-18
中文摘要: 差别矩阵属性约简是粗糙集重要约简方法之一,但在处理不一致大数据集时存在不足。为此,首先提出决策差别集的概念,并给出基于决策差别集的属性约简定义,同时研究了由该定义获得的约简与正区域约简之间的等价性。接着,给出水平划分决策表的方法,并将子决策表分配到不同的网络节点上构建子决策差别集,并行完成核属性和属性约简求解。实例分析和UCI中数据集的实验比较表明所提出的约简算法是正确的、高效的。
中文关键词: 粗糙集  决策差别集  核属性  属性约简
Abstract:The notion of decision discernibility set and definition of attribute reduction based on decision discernibility set were presented.It was proved that attribute reduction acquired from the definition is equivalence to attribute reduction based on positive region.And then, the method of horizontally partitioning decision table was proposed and the sub-decision table can be assigned to different network nodes and finish computing core attribute and attribute reduction based on sub-decision discernibility set.Finally, the example analysis experiment results form datasets of UCI showed that the proposed parallel algorithms are efficient and effective.
文章编号:201301271     中图分类号:    文献标志码:
基金项目:国家自然科学基金资助项目(5130711);安徽省自然科学基金资助项目(1308085QF114);安徽高等学校省级自然科学研究重点资助项目(KJ2013A015;KJ2012A212);滁州学院优秀青年人才基金重点项目(2013RC003);计算智能与信号处理教育部重点实验室开发课题基金资助项目
作者简介:
引用文本:
葛浩,李龙澍,徐怡,杨传健.水平划分决策表的属性约简算法[J].工程科学与技术,2014,46(3):89-94.
Ge Hao,Li Longshu,Xu Yi,Yang Chuanjian.An Algorithm for Attribute Reduction Based on Horizontally Partitioning Decision Table[J].Advanced Engineering Sciences,2014,46(3):89-94.