[1]吴婷婷,李 艳,郭娜娜,等.基于优势-等价关系的属性约简算法[J].南京师范大学学报(自然科学版),2017,40(03):45.[doi:10.3969/j.issn.1001-4616.2017.03.007]
　Wu Tingting,Li Yan,Guo Nana,et al.Attribute Reduction Algorithms Based on Dominance-Equivalence Relations[J].Journal of Nanjing Normal University(Natural Science Edition),2017,40(03):45.[doi:10.3969/j.issn.1001-4616.2017.03.007]
点击复制

基于优势-等价关系的属性约简算法()
分享到：

《南京师范大学学报》（自然科学版）[ISSN:1001-4616/CN:32-1239/N]

卷:
第40卷
期数:
2017年03期
页码:
45
栏目:
·计算机科学·
出版日期:
2017-09-30

文章信息/Info

Title:
Attribute Reduction Algorithms Based on Dominance-Equivalence Relations
文章编号:
1001-4616(2017)03-0045-07
作者:
吴婷婷1; 李 艳1; 郭娜娜1; 何 强2
(1.河北大学数学与信息科学学院,河北省机器学习与计算智能重点实验室,河北 保定 071002)(2.北京建筑大学理学院,北京 100044)
Author(s):
Wu Tingting1; Li Yan1; Guo Nana1; He Qiang2
(1.College of Mathematics and Information Science,Hebei University,Key Lab. of Machine Learning and Computational Intelligence,Baoding 071002,China)(2.College of Science,Beijing University of Civil Engineering and Architecture,Beijing 100044,China)
关键词:
粗糙集; 优势-等价关系; 属性约简; 辨识矩阵; 样例对
Keywords:
rough set; dominance-equivalence relation; attributes reduction; discernibility matrix; sample pair
分类号:
TP311
DOI:
10.3969/j.issn.1001-4616.2017.03.007
文献标志码:
A
摘要:
考虑多标准分类问题,即条件属性具有偏好关系而决策属性是无序的类别,通过在条件属性上引入优势关系而决策属性仍然用等价关系来描述不同的属性. 针对这类信息系统,本文提出了一种基于样例对的矩阵约简算法. 区别于传统的基于辨识矩阵约简方法,该算法在不计算辨识矩阵的前提下,通过选择样例对,来找到辨识矩阵中对约简有用的属性,因此,所提算法能够明显改善计算约简的时间耗费. 进一步,为了处理较大规模的数据,提出了一种近似约简算法,该算法按属性重要性添加属性到约简中,进一步缩短了求取约简的时间. 最后在UCI数据集上进行大量的实验与传统的约简算法进行了对比,表明了所提出算法的可行性与有效性.
Abstract:
Considering multiple criteria classification problems,dominance relations and equivalence relations can be respectively introduced to condition attributes and decision attributes to describe different types of data. Based on the dominance-equivalence relations,a novel attribute reduction method based on sample pair selection is developed to deal with this kind of information systems. Instead of calculating the whole discernibility matrix,the proposed method only store the useful attributes for attribute reduction by selecting the discerned sample pairs,and therefore it can significantly improve the time costin attribute reduction. In addition,we propose an approximate reduction algorithm in order to deal with comparative large-scale information systems. This algorithm add attributes based on attribute importance and it’s time saving. Finally,the experimental results on UCI data sets demonstrate the feasibility and effectiveness of the proposed algorithms.

参考文献/References:

[1] PAWLAK Z. Rough sets[J]. International journal of information and computer sciences,1982,11(3):341-356.
[2]PAWLAK Z. Rough sets:theoretical aspects of reasoning about data[M]. Boston:Kluwer Academic Publishers,1991.
[3]苗夺谦,李道国. 粗糙集理论、算法与应用[M]. 北京:清华大学出版社,2008.
[4]ZHANG Q,XIE Q,WANG G. A survey on rough set theory and its applications[J]. CAAI transactions on intelligence technology,2016,1(4):323-333.
[5]YAO J T,LINGRAS P,WU W Z,et al. Rough sets and knowledge technology[C]//Proceeding of the 2nd Canadian. Toronto:Rough Sets Technology,2007.
[6]CHENG Q,QI Z,ZHANG G,et al. Robust modeling and prediction of thermally induced positional error based on grey rough set theory and neural networks[J]. The international journal of advanced manufacturing technology,2016,83(5):1-12.
[7]PéREZ D N,RUANO O D,FDEZ R F,et al. Boosting accuracy of classical machine learning antispam classifiers in real scenarios by applying rough set theory[J]. Scientific programming,2016,2016(3/4):1-10.
[8]PENG L,NIU R,HUANG B,et al. Landslide susceptibility mapping based on rough set theory and support vector machines:a case of the three gorges area,China[J]. Geomorphology,2014,204(1):287-301.
[9]王珏,苗夺谦,周育健. 关于Rough Set 理论与应用的综述[J]. 模式识别与人工智能,1996,9(4):337-344.
[10]张文修,梁怡,吴伟志. 信息系统与知识发现[M]. 北京:科学出版社,2003.
[11]张文修,米据生,吴伟志. 不协调目标信息系统的知识约简[J]. 计算机学报,2003,26(1):12-18.
[12]KRYSZKIEWICZ M. Comparative studies of alternative of knowledge reduction in inconsistent systems[J]. Intelligent systems,2001,16(1):105-120.
[13]GRECOS,MATARAZZO B,SLOWINSKI R. Rough approximation of preference relation by dominance relations[J]. European journal of operational research,1999,117(1):63-83.
[14]徐伟华,张文修. 基于优势关系下不协调目标信息系统的知识约简[J]. 计算机科学,2006,33(2):182-184.
[15]王国胤. Rough集理论与知识获取[M]. 西安:西安交通大学出版社,2001.
[16]STEFANOWSKI J. Handling continuous attributes in discovery of strong decision rules[M]. Berlin:Springer Berlin Heidelberg,1998.
[17]SAI Y,YAO Y Y,ZHANG N. Data analysis and mining in order information table[C]//IEEE International Conference on Data Mining. USA:IEEE Computer Society Press,2001:497-504.
[18]SHAO M W,ZHANG W X. Dominance relation and rules in an incomplete ordered information system[J]. International journal of intelligent systems,2005,20:13-27.
[19]张文修,姚一豫,梁怡. 粗糙集与概念格[M]. 西安:西安交通大学出版社,2006.
[20]苟光磊,王国胤. 基于不协调置信优势原理关系的知识约简[J]. 计算机科学,2016,43(6):204-207.
[21]黄琴,魏玲. 基于布尔矩阵的序信息系统属性约简方法[J]. 小型微型计算机系统,2016,37(8):1 717-1 720.
[22]贺明利,魏玲. 基于优势关系的序形式背景约简[J]. 计算机科学,2015,42(6):46-49.
[23]CHEN D G,ZHAO S Y,ZHANG L,et al. Sample pair selection for attribute reduction with rough set[J]. IEEE Trans Know Data Eng,2012,24(1):2 080-2 093.

相似文献/References:

[1]吴国兵,黄兵.粗糙群定义的缺陷与修正[J].南京师范大学学报(自然科学版),2011,34(04):39.
　Wu Guobing,Huang Bing.Defect and Revision of Definition of Rough Group[J].Journal of Nanjing Normal University(Natural Science Edition),2011,34(03):39.
[2]陆 悠,华 泽,奚雪峰,等.一种面向增删操作的粗糙集属性约简更新算法[J].南京师范大学学报(自然科学版),2015,38(01):48.
　Lu You,Hua Ze,Xi Xuefeng,et al.An Attribute Reduction Update Algorithm for Object’sAdding-Deleting Based on Rough Set Theory[J].Journal of Nanjing Normal University(Natural Science Edition),2015,38(03):48.
[3]张 静,鞠恒荣,杨习贝,等.柔性多粒化决策理论粗糙集模型[J].南京师范大学学报(自然科学版),2017,40(01):48.[doi:10.3969/j.issn.1001-4616.2017.01.008]
　Zhang Jing,Ju Hengrong,Yang Xibei,et al.A Flexible Multigranulation Decision-Theoretic Rough Set Model[J].Journal of Nanjing Normal University(Natural Science Edition),2017,40(03):48.[doi:10.3969/j.issn.1001-4616.2017.01.008]

备注/Memo

备注/Memo:
收稿日期:2017-03-18.
基金项目:国家自然科学基金(61170040、61473111)、河北省自然科学基金(F2014201100、 A2014201003).
通讯联系人:李艳,博士,教授,研究方向:机器学习. E-mail:ly@hbu.cn
更新日期/Last Update: 2017-09-30