(1)若λ1i<λ,则
然而,
综上可知定理6成立.
证毕.
定理6说明当λ在区间[0.5,1]上以及粗糙集的边界域被细分时,近似集的模糊度也会呈单调性变化,因此也可设计基于近似集模糊度的属性约简方法.
5 基于模糊度的属性约简算法
定义8(决策信息系统的近似模糊度)设S为一个决策信息系统,U为论域,C为条件属性集,D为决策属性集,B⊆C,U/IND(B)={X1,X2,…,Xn}为条件属性集B在U上导出的划分,U/IND(D)={Y1,Y2,…,Ym}为决策属性集D在U上导出的划分,称
(3)
为决策信息系统S的近似模糊度.
显然式(3)也满足定义6中的模糊度公式.
定理7在决策信息系统S中,U为论域,C为条件属性集,D为决策属性集,B⊆C,U/IND(B)={X1,X2,…,Xn}为条件属性集B在U上导出的划分,U/IND(D)={Y1,Y2,…,Ym}为决策属性集D在U上导出的划分.若U/Bi+1U/Bi,则d0.5(DBi+1)≤d0.5(DBi).
显故
即d0.5(DBi+1)≤d0.5(DBi)成立.
证毕.
根据定理7我们可设计基于粗糙集近似集模糊度的属性约简算法.
该算法的时间复杂度主要来源于Step1和Step3.在Step1中的时间复杂度为O(|U|2),Step3中的时间复杂度为O(|U|3),因此该算法总的时间复杂度为O((|U|2)+(|U|3)).这与基于条件信息熵的属性约简算法的时间复杂度相同.我们将在后继的研究中,试图把该结果用到图像的边缘分割中,建立增量式特征选择和属性约简算法等,以希望提高图像边缘分割算法的精确性和容错性.
6 结束语
当前科技的不断发展让不确定性信息的研究变得日渐重要[20,21].粗糙集作为一种能有效处理不确定性信息的理论,其不确定性度量也是目前研究的一个重要内容[22,23].本文主要分析了相似度在多粒度知识空间下的变化规律,定义了近似集的模糊度,同时也分析了模糊度在多粒度知识空间下的变化规律.该研究为用近似集方法进行属性约简提供了理论基础.希望这些工作能够推动不确定信息理论的发展,扩展粗糙集理论模型,促进其应用.
[1]Pawlak Z.Rough sets[J].Information Journal of Computer and Information Science,1982,11(5):341-365.
[2]冯林,王国胤,李天瑞.连续值属性决策表中的知识获取方法[J].电子学报,2009,37(11):2432-2438.
Feng Lin,Wang Guoyin,Li Tianrui.Knowledge acquisition from decision tables containing continuous-valued attributes[J].Acta Electronica Sinica,2009,37(11):2432-2438.(in Chinese)
[3]张腾飞,肖健梅,王锡淮.粗糙集理论中属性相对约简算法[J].电子学报,2005,33(11):2080-2083.
Zhang Tengfei,Xiao Jianmei,Wang Xihuai.Algorithms of attribute relative reduction in rough set theory[J].Acta Electronica Sinica,2005,33(11):2080-2083.(in Chinese)
[4]Inuiguchi M,Yoshioka Y,Kusunoki Y.Variable-precision dominance-based rough set approach and attribute reduction[J].International Journal of Approximate Reasoning,2009,50(8):1199-1214.
[5]Xie G,Zhang J,Lai K K,et al.Variable precision rough set for group decision-making;An application[J].International Journal of Approximate Reasoning,2008,49(2):331-343.
[6]Liu J N K,Hu Y,He Y.A set covering based approach to find the reduct of variable precision rough set[J].Information Sciences,2014,275:83-100.
[7]Mi J S,Wu W Z,Zhang W X.Approaches to knowledge reduction based on variable precision rough set model[J].Information Sciences,2004,159(3):255-272.
[8]Qian Y,Zhang H,Sang Y,et al.Multigranulation decision-theoretic rough sets[J].International Journal of Approximate Reasoning,2014,55(1):225-237.
[9]张仕光,米据生,胡清华.粗糙ε-支持向量回归模型[J].南京大学学报 (自然科学版),2013,49(5):650-654.
Zhang Siguang,Mi Jusheng,Hu Qinghua.Roughε-support vector regression model[J].Journal of Nanjing University (Natural Science),2013,49(5):650-654.(in Chinese)
[10]李顺勇,钱宇华.基于多粒度粗糙决策下的属性约简算法[J].中北大学学报(自然科学版),2013,34(5):589-592.
Li Shunyong,Qian Yuhua.Attribute reduction algorithm based on multi-granularity rough decision[J].Journal of North University of China (Natural Science),2013,34(5):589-592.(in Chinese)
[11]Yao Y.Two semantic issues in a probabilistic rough set model[J].Fundamenta Informaticae,2011,108(3):249-265.
[12]Yao Y.Probabilistic rough set approximations[J].International Journal of Approximate Reasoning,2008,49(2):255-271.
[13]Ziarko W.Probabilistic approach to rough sets[J].International Journal of Approximate Reasoning,2008,49(2):272-284.
[14]张清华,王国胤,肖雨.粗糙集的近似集[J].软件学报,2012,23(7):1745-1759.
Zhang Qinghua,Wang Guoyin,Xiao Yu.Approximation sets and rough set[J].Journal of Software,2012,23(7):1745-1759.(in Chinese)
[15]Zhang Q,Guo Y,Xiao Y.Attribute reduction based on approximation set of rough set[J].Journal of Computational Information Systems,2014,10(16):6859-6866.
[16]Zhang Q,Wang J,Wang G,et al.The approximation set of a vague set in rough approximation space[J].Information Sciences,2015,300:1-19.
[17]王国胤,张清华.不同知识粒度下粗糙集的不确定性研究[J].计算机学报,2008,31(9):1588-1598.
Wang Guoyin,Zhang Qinghua.Uncertainty of rough sets in different knowledge granularities[J].Chinese Journal of Computers,2008,31(9):1588-1598.(in Chinese)
[18]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.
Wang Guoyin.Rough Set Theory and Knowledge Discovery[M].Xi’an:Xi’an Jiaotong University Press,2001.(in Chinese)
[19]杨纶标,高英仪,凌卫新.模糊数学原理及应用[M].广州:华南理工大学出版社,2011.
Yang Lunbiao,Gao Yingyi,Ling Weixin.Principles and Applications of Fuzzy Mathematics[M].Guangzhou:South China University of Technology Press,2011.(in Chinese)
[20]苗夺谦,王国胤,刘清,等.粒计算:过去,现在与展望[M].科学出版社,2007.
Miao Duoqian,Wang Guoyin,Liu Qing,et al.Granular Computing:Past,Present and Future Prospects[M].Beijing:Science Press,2007.(in Chinese)
[21]Zhang Q,Wang J,Wang G,Hu F.Approximation set of the interval set in pawlak’s space[J].The Scientific World Journal,2014,Article ID 317387,1-12.
[22]杨明.决策表中基于条件信息熵的近似约简[J].电子学报,2007,35(11):2156-2160.
Yang Ming.Approximate reduction based on conditional information entropy in decision tables[J].Acta Electronica Sinica,2007,35(11):2156-2160.(in Chinese)
[23]Sun B Z,Ma W M.Uncertainty measure for general relation-based rough fuzzy set[J].Kybernetes,2013,42(6):979-992.
张清华男,1974年11月出生,重庆沙坪坝人.教授,硕士生导师.1998年、2003年和2009年分别在四川大学、重庆邮电大学和西南交通大学获理学学士、工学硕士和工学博士学位.现为中国人工智能学会粗糙集与软计算专委会秘书长,主要从事不确定信息处理、粗糙集与粒计算等方面的研究工作.
E-mail:zhangqh@cqupt.edu.cn
薛玉斌男,1990年5月出生,重庆潼南人.硕士研究生,研究方向为不确定信息处理、粗糙集与粒计算.
Research on Uncertainty of Approximation Set of Rough Set
ZHANG Qing-hua1,2,XUE Yu-bin1,HU Feng2,YU Hong2
(1.School of Science,Chongqing University of Posts and Telecommunications,Chongqing 400065,China; 2.Chongqing Key Laboratory of Computational Intelligence,Chongqing University of Posts and Telecommunications,Chongqing 400065,China)
Rough set describes an uncertain target set with upper and lower approximation sets,and approximation set of rough set uses 0.5-approximation set as an approximation set of the uncertain target set.In this paper,we firstly find that the theory of attribute reduction algorithm based on similarity between target set and its 0.5-approximation set is still incomplete,and this similarity is not sensitive to changing granularities.In order to overcome above shortcomings,the change rule of similarity with changing granularities in a multi-granularity space is analyzed,fuzzy degree of approximation set is defined,and the change rules of this fuzziness with changing granularities are analyzed in detail in a hierarchical space.Finally,a new attribute reduction algorithm is proposed.From a new perspective,a kind of differentiation measure between an uncertain target set and its approximation set is presented.
rough set;approximation set;fuzziness;uncertainty;multi-granularity
2014-12-17;
2015-03-22;责任编辑:李勇锋
国家自然科学基金(No.61472056,No.61309014,No.61379114);重庆市自然科学基金(No.cstc2012jjA40032,cstc2013jcyjA4006)
TN911.23
A
0372-2112 (2016)07-1574-07
��学报URL:http://www.ejournal.org.cn
10.3969/j.issn.0372-2112.2016.07.008