摘要
序贯三支决策方法是一种能够表示问题中的多重层次粒度,并将多粒度结合起来解决不确定决策问题的有效途径。优势-等价关系粗糙集则是针对条件属性具有偏好关系的分类问题,提取有序信息,对目标概念进行近似,从而形成决策知识。利用传统的优势关系粗糙集方法进行知识约简和提取的效率低下,而目前大部分序贯三支决策方法则局限在符号值属性的信息系统中,对连续值和有序值不能进行有效处理,造成一定程度的信息丢失。因此,将序贯三支决策的思想应用于优势关系粗糙集模型中,定义了一种新的基于序贯三支决策的属性约简及相应的属性重要度,对具有偏好值属性的信息系统进行更加高效的处理,通过多粒度的表示和关系的研究,加速了知识约简过程。选取了多组UCI数据进行实验,结果表明所提出的基于优势关系的序贯三支决策方法能够在保证约简质量的基础上明显降低时间耗费。
Sequential three-way decision is an effective way to solve problems under multiple levels granularity.Dominance-equivalence relation based rough set approach can be used to handle classification problems for conditional attributes with preference ordered,extract related information,approximate target concepts and finally form the decisionmaking knowledge.The traditional dominance relation-based rough sets model is very time consuming for knowledge reduction and extraction,however,most of current sequential three-way decision models are limited to information systems of symbolic attributes,which can not process continuous and ordinal values effectively,and will cause a certain degree loss of information.Therefore,this paper applied the idea of sequential three-way decisions to the dominance relation-based rough sets models,defined a new attribute reduction method based on sequential three-way decisions and the corresponding attribute importance measure,and then accelerated the processing of information systems with ordinal attributes.Finally,the efficiency of knowledge reduction is improved through multiple granularity representations and relationships.Several UCI data sets are selected for experiments.The results show that the proposed sequential three-decision method based on dominance relations can reduce the time consumption noticeably and guarantee the quality of the attribute reduction.
引文
[1] GRECO S,MATARAZZO B,SLOWINSKI R.Rough approximation by dominance relations[J].International Journal of Intelligent Systems,2002,17(2):153-171.
[2] CHEN J,WANG G Y,HU J.Positive domain reduction based on dominance relation in inconsistent system[J].Computer Science,2008,35(3):216-218.(in Chinese)陈娟,王国胤,胡军.优势关系下不协调信息系统的正域约简[J].计算机科学,2008,35(3):216-218.
[3] LI Y,SUN N X,ZHAO J,et al.Reductions based on dominanceequivalence relations and rule extraction methods[J].Computer Science,2011,38(11):220-224.(in Chinese)李艳,孙娜欣,赵津,等.基于优势-等价关系的几种约简及规则抽取方法[J].计算机科学,2011,38(11):220-224.
[4] JIN Y,LI Y,HE Q.A fast positive-region reduction method based on dominance-equivalence relations[C]∥International Conference on Machine Learning and Cybernetics.IEEE,2017:152-157.
[5] AWANG M I,ROSE A N M,AWEANG M K,et al.Multiple criteria preference relation by dominance relations in soft set theory[M].Berlin:Springer International Publishing,2016:475-484.
[6] YAO Y Y.Decision-theoretic rough set models[C]∥International Conference on Rough Sets and Knowledge Technology.Springer,Berlin,Heidelberg,2007:1-12.
[7] YAO Y Y,WONG S K M.A decision theoretic framework for approximating concepts[J].International Journal of Man-Machine Studies,1992,37(6):793-809.
[8] YAO Y Y,ZHAO Y.Attribute reduction in decision-theoretic rough set models[J].Information Sciences,2008,178(17):3356-3373.
[9] YAO Y Y.An outline of a theory of three-way decisions[C]∥International Conference on Rough Sets and Current Trends in Computing.Springer,Berlin:Heidelberg,2012:1-17.
[10]贾修一,商林,周献中,等.三支决策理论与应用[M].南京:南京大学出版社,2012.
[11]YAO Y.Rough sets and three-way decisions[M].Berlin:Springer International Publishing,2015.
[12]LIU D,YAO Y Y,LI T R.Three-way investment decisions with decision-theoretic rough sets[J].International Journal of Computational Intelligence Systems,2011,4(1):66-74.
[13]LINGRAS P,CHEN M,MIAO D.Rough cluster quality index based on decision theory[J].IEEE Transactions on Knowledge and Data Engineering,2009,21(7):1014-1026.
[14]YU H,LIU Z,WANG G.An automatic method to determine the number of clusters using decision-theoretic rough set[M].Amsterdam:Elsevier Science Inc.2014.
[15]LI H,ZHANG L,HUANG B,et al.Sequential three-way decision and granulation for cost-sensitive face recognition[J].Knowledge-Based Systems,2016,91(C):241-251.
[16]JIA X,SHANG L,ZHOU B,et al.Generalized attribute reduct in rough set theory[J].Knowledge-Based Systems,2016,91(C):204-218.
[17]MENG Z,SHI Z.On quick attribute reduction in decision-theoretic rough set models[J].Information Sciences,2016,330:226-244.
[18]LI J,HUANG C,QI J,et al.Three-way cognitive concept learning via multi-granularity[J].Information Sciences,2017,378(1):244-263.
[19]LI J,MEI C,XU W,et al.Concept learning via granular computing:a cognitive viewpoint[J].Information Sciences,2015,298(1):447-467.
[20]PAWLAK Z.Rough sets:theoretical aspects of reasoning about data[M].Boston:Kluwer Academic Publishers,1991.
[21]QIAN J,DANG C,YUE X,et al.Attribute reduction for sequential three-way decisions under dynamic granulation[J].International Journal of Approximate Reasoning,2017,85:196-216.
[22]XU W H,ZHANG X X,ZHANG W X.Upper approximation reduction in inconsistent target information system based on dominance relations[J].Computer Engineering,2009,35(18):191-193.(in Chinese)徐伟华,张晓燕,张文修.优势关系下不协调目标信息系统的上近似约简[J].计算机工程,2009,35(18):191-193.
[23]BACHE K,LICHMA M.UCI Machine Learning Repository[OL].http://archive.ics.uci.edu/ml.