[1]李 冬,蒋 瑜*,鲍杨婉莹.基于属性质量度的变精度邻域粗糙集属性约简[J].四川师范大学学报(自然科学版),2020,43(04):560-568.[doi:10.3969/j.issn.1001-8395.2020.04.022]
 LI Dong,JIANG Yu,BAOYANG Wanying.Attribute Reduction of Variable Precision Neighborhood Rough Sets Based on Attribute Quality[J].Journal of SichuanNormal University,2020,43(04):560-568.[doi:10.3969/j.issn.1001-8395.2020.04.022]
点击复制

基于属性质量度的变精度邻域粗糙集属性约简()
分享到:

《四川师范大学学报(自然科学版)》[ISSN:1001-8395/CN:51-1295/N]

卷:
43卷
期数:
2020年04期
页码:
560-568
栏目:
基础理论
出版日期:
2020-06-20

文章信息/Info

Title:
Attribute Reduction of Variable Precision Neighborhood Rough Sets Based on Attribute Quality
文章编号:
1001-8395(2020)04-0560-09
作者:
李 冬 蒋 瑜* 鲍杨婉莹
成都信息工程大学 软件工程学院, 四川 成都 610000
Author(s):
LI Dong JIANG Yu BAOYANG Wanying
School of Software Engineering, Chengdu University of Information Technology, Chengdu 610000, Sichuan
关键词:
变精度邻域粗糙集 属性约简 正域 属性质量度 正确分类率
Keywords:
variable precision neighborhood rough set attribute reduction positive domain attribute quality correct classification rate
分类号:
TP18; O159
DOI:
10.3969/j.issn.1001-8395.2020.04.022
文献标志码:
A
摘要:
变精度邻域粗糙集相比于邻域粗糙集具有抗噪容错的能力,但由于重新定义了下近似,正域的划分不再严格,使得属性重要度的可信度降低,在精度改变的情况下无法优先选取最优的属性.针对这一问题,分析变精度邻域粗糙集的下近似,引入邻域内的正确分类率,定义属性质量度,提出一种基于正域的增量和平均正确分类率的增率相结合的属性度量方法.通过和现有的基于属性重要度的属性约简算法做比较,实验结果表明,改进后的属性度量方法对变精度有更好的适应性,在不同变精度阈值下能得到更优的约简结果.
Abstract:
Compared with neighborhood rough sets, variable precision neighborhood rough sets have the ability of anti-noise and fault-tolerant. However, since the lower approximation is redefined and the division of the positive domain is no longer strict, the reliability of the attribute importance is reduced, and the optimal attribute cannot be prioritized in the case of a change in accuracy. In order to solve this problem, this paper analyzes the lower approximation of the variable precision neighborhood rough set, introduces the correct classification rate in the neighborhood, defines an attribute quality, and proposes an attribute measurement method based on the combination of the positive domain increment and the rate of increase in the average correct classification rate. By comparing with the existing attribute reduction algorithm based on attribute importance, the experimental results show that the improved attribute measurement method has better adaptability to variable precision, and better prediction results can be obtained under different variable precision thresholds.

参考文献/References:

[1] PAWLAK Z. Rough set[J]. International J Computer and Information Sciences,1982,11(5):341-356.
[2] 石云,孙玉芳,左春. 基于Rough Set的空间数据分类方法[J]. 软件学报,2000(5):673-678.
[3] TAY F E H, SHEN L. Economic and financial prediction using rough sets model[J]. European J Operational Research,2002,141(3):641-659.
[4] XIONG X, CHEN L, LIANG J. Analysis of roadway traffic accidents based on rough sets and bayesian networks[J]. Promet-Traffic & Transportation,2018,30(1):71-81.
[5] 李伟岸. 基于粗糙集理论和神经网络方法的飞行冲突解脱研究[D]. 天津:中国民航大学,2018.
[6] 李兵洋,肖健梅,王锡淮. 多半径邻域粗糙集改进约简算法[J]. 计算机工程与应用,2017,53(11):7-12.
[7] ZADEH L A. Toward a theory of fuzzy information granulation and its centrality in human reasoning and fuzzy logic[J]. Fuzzy Sets & Systems,1997,90(90):111-127.
[8] LIN T Y. Granular computing on binary relations I:data mining and neighborhood systems[J]. Rough Sets in Knowledge Discovery,1998,18(1):107-121.
[9] HU Q, YU D, LIU J, et al. Neighborhood rough set based heterogeneous feature subset selection[J]. Information Sciences,2008,178(18):3577-3594.
[10] 胡清华,赵辉,于达仁. 基于邻域粗糙集的符号与数值属性快速约简算法[J]. 模式识别与人工智能,2008,21(6):732-738.
[11] 王鹏. 基于邻域粗糙集的属性约简算法研究[D]. 石家庄:河北科技大学,2011.
[12] YING Y, PEDRYCZ W, MIAO D. Neighborhood rough sets based multi-label classification for automatic image annotation[J]. International J Approximate Reasoning,2013,54(9):1373-1387.
[13] YONG L, HUANG W, JIANG Y, et al. Quick attribute reduct algorithm for neighborhood rough set model[J]. Information Sciences,2014,271:65-81.
[14] CHEN H M, LI T R, CAI Y, et al. Parallel attribute reduction in dominance-based neighborhood rough set[J]. Information Sciences,2016,373:351-368.
[15] ZIARKO W. Variable precision rough set model[J]. J Computer System Science,1993,46(1):39-59.
[16] 贾俊芳,张英. 基于属性重要度的变精度邻域粗糙集知识约简[J]. 山西大同大学学报(自然科学版),2014,30(6):1-3.
[17] RUOMING A N, MINGLIANG S, ASTRONAUTICS S O. Application of attributes reduction and weights calculation through neighborhood rough set[J]. Computer Engineering & Applications,2016,52(7):169-165.
[18] 彭潇然,刘遵仁,纪俊. 基于样本类别的邻域粗糙集正域计算[J]. 计算机工程与应用,2018,54(20):115-121.

相似文献/References:

[1]舒畅.多值RS理论中属性约简的另一种算法[J].四川师范大学学报(自然科学版),2005,(01):43.
 SHU Chang(College of Mathematics and Software Science,Sichuan Normal University,Chengdu 00,et al.[J].Journal of SichuanNormal University,2005,(04):43.
[2]胡丹,莫智文.关于粗糙集理论与信息熵的几点注记[J].四川师范大学学报(自然科学版),2002,(03):257.
 HU Dan,MO Zhi-wen (College of Mathematics and Software Science,Sichuan Normal University,et al.[J].Journal of SichuanNormal University,2002,(04):257.
[3]王燕,莫智文,唐孝,等.基于集对分析的多值信息系统粗集模型及其约简[J].四川师范大学学报(自然科学版),2007,(03):318.
 WANG Yan,MO Zhi-wen,TANG Xiao,et al.[J].Journal of SichuanNormal University,2007,(04):318.
[4]韩 楠,莫智文*,舒 畅.不完备贝叶斯决策信息系统的属性约简[J].四川师范大学学报(自然科学版),2016,(06):825.[doi:10.3969/j.issn.1001-8395.2016.06.008]
 HAN Nan,MO Zhiwen,SHU Chang.Attribute Reduction in Incomplete Bayesian Decision Information System[J].Journal of SichuanNormal University,2016,(04):825.[doi:10.3969/j.issn.1001-8395.2016.06.008]

备注/Memo

备注/Memo:
收稿日期:2019-03-11 接受日期:2019-05-21
基金项目:四川省教育厅重点项目(17ZA0071)
*通信作者简介:蒋 瑜(1980—),男,副教授,主要从事粗糙集、数据挖掘与智能计算的研究,E-mail:jiangyu@cuit.edu.cn
更新日期/Last Update: 2020-06-20