基于三支决策的模糊信息系统OWA算子参数选择  

Three-way Decisions Based Parameter Selection of OWA Operators in Fuzzy Information System

作　　者：杨霁琳[1] 张贤勇[2] 唐孝[2] 

机构地区：[1]四川师范大学基础教学学院,成都610068 [2]四川师范大学数学与软件科学学院,成都610068

出　　处：《数据采集与处理》2016年第6期1156-1163,共8页Journal of Data Acquisition and Processing

基　　金：国家自然科学基金(61203285;61303204;61673285)资助项目;四川省教育厅科研(15ZB0028)资助项目

摘　　要：在模糊信息系统中,通过有序加权平均(Ordered weighted averaging,OWA)算子计算对象相似度,可以建立λ截集的相容关系。当λ的值确定后,OWA算子中量词参数(α,β)的选取直接关系到不可区分关系的建立以及信息粒的粗细。因此合理选取参数(α,β)是值得进一步研究的问题。本文采用粗糙集三支决策理论,研究模糊信息系统OWA算子参数选择的相关内容。基于参数的激进、中庸和消极语义,确定OWA算子(α,β)的3种常用取值;进而研究相似度、相容类、双向近似和三支区域在3种参数选择下的性质关系。最后利用实例验证分析了模糊量词参数语义解释的合理性。本文采用三支决策创新视角,得到模糊信息系统OWA算子的深入性质,为相关模糊量词参数提供语义解释与选择依据。In the fuzzy information system ,tolerance relation of the λlevel sets based on the similar de‐gree of the objects can be obtained by the ordered weighted averaging （OWA ） operator .Indiscernibility relation and grain size are affected by the fuzzy quantifier parameters （α,β） of the OWA operators ,when the value of λis the same .Therefore ,it is worth to study how to reasonably select fuzzy quantifier pa‐rameter （α,β） values .Parameter selection of OWA operators is discussed based on three‐way decisions of rough set theory in fuzzy information system .According to radicalness ,mean ,and negative of parame‐ters semantics ,three most commonly used values of （α,β） are proposed .Furthermore ,related properties of similar degrees ,tolerance classes ,upper and lower approximation sets as well as three‐way regions are discussed based on three most commonly used values of （α,β） .Finally ,experimental results are used to analyze reasonableness of semantic explanations about fuzzy quantifier parameters .The study adopts a novel viewpoint of three‐way decisions theory ,and the new semantic explanations and validity of selection methods about related fuzzy quantifier parameters （α,β） in the OWA operators are given in fuzzy infor‐mation system .

关 键 词：模糊信息系统 粗糙集 三支决策 OWA算子 相容关系 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]