基于属性重要度并行约简算法的优化

被引:6
作者
陈林
邓大勇
闫电勋
机构
[1] 浙江师范大学数理信息学院
关键词
粗糙集; 属性重要度; 矩阵; 并行约简;
D O I
10.13232/j.cnki.jnju.2012.04.014
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
针对约简长度和时间效率两方面问题,提出了一种基于属性重要度并行约简的优化算法.该算法通过对每个子表赋权值,并在所建立的属性重要度矩阵中选择权值之和最大的列所对应的属性作为约简属性,所得到的约简即为并行约简.最后,通过UCI机器学习数据库中的几个实例验证了改进后算法的正确性和有效性.
引用
收藏
页码:376 / 382
页数:7
相关论文
共 14 条
[1]  
Attribute reduction among decision tables by voting. Deng D Y. Proceedings of2008IEEE In-ternational Conference of Granular Computing . 2008
[2]  
Finding reducts in composed information systems. Kryszkiewicz M,Rybinski H. Proceedings of the Second International Workshop on Rough Sets and Knowledge Discovery (RSKD’’93) . 1993
[3]   Reduction Algorithms Based on Discernibility Matrix:The Ordered Attributes Method [J].
王珏 ;
王驹 .
Journal of Computer Science and Technology, 2001, (06) :489-504
[4]   属性最小约简的增量式算法 [J].
刘宗田 .
电子学报, 1999, (11) :97-99
[5]  
(F,ε)-Parallel reducts in a series of decision subsystems. Deng D Y. Proceedings of the3 rd In-ternational Joint Conference on Computational Sciences and Optimization . 2010
[6]  
http://archive.ics.uci.edu/ml/ .
[7]   不完备信息系统中Rough集的扩充模型 [J].
尹旭日 ;
商琳 .
南京大学学报(自然科学版), 2006, (04) :337-341
[8]  
A Comparison of Dynamic and Non-Dynamic Rough Set Methods for Extracting Laws from Decision Tables. Bazan JG. Rough Sets in Knowledge Discovery 1: Methodology and Applications . 1998
[9]  
Rough sets: Theoretical aspects of reasoning about data. Pawlak Z. . 1991
[10]  
Comparison of parallel reducts and dynamic reducts in theory. Deng D Y. Computer Science . 2009