改进的区间值决策表决策代价最小化属性约简  

An Improved Interval-valued Decision Table Decision Cost Minimization Attribute Reduction

作　　者：梁艳玲 唐孝 周施吉[1,2] LIANG Yan-ling;TANG Xiao;ZHOU Shi-ji(School of Mathematical Sciences,Sichuan Normal University,Chendu 610066,China;Institute of Intelligent Information and Quantum Information,Sichuan Normal University,Chengdu 610066,China)

机构地区：[1]四川师范大学数学科学学院,四川成都610066 [2]四川师范大学智能信息与量子研究信息所,四川成都610066

出　　处：《模糊系统与数学》2024年第1期131-144,共14页Fuzzy Systems and Mathematics

基　　金：国家自然科学基金(61673285);四川省科技计划资助项目(2019YJ0529);四川省教育厅重点项目(18ZA0410)。

摘　　要：近年来,基于贝叶斯决策代价最小化的属性约简在三支决策区间值系统中引起了众多学者的关注.然而在构造损失函数时,少有研究考虑由于相容等价类之间的差异性对约简结果带来的影响,针对这一问题,提出一种基于距离度量的决策代价最小化属性约简方法.文章首先通过度量决策类与相容类之间的距离,建立综合损失函数来描述对象之间的差异信息;然后根据贝叶斯决策规则和预先给定的初始代价函数,构建了自适应閾值对的计算方法,建立了区间值决策系统中决策成本最小化的属性约简算法;最后给出算例,结果表明所提出的约简算法能获得决策成本更小的约简结果,且在理论上更具有合理性。In recent years,attribute reduction based on Bayesian decision cost minimization has attracted the attention of many scholars in the three-way decisions interval-valued systems.However,in the construction of loss function,few studies consider the influence of the difference between the compatible equivalence classes on the reduction result.To solve this problem,a attribute reduction method of decision cost minimization based on distance measure is proposed.By measuring the distance between decision class and compatible class,a comprehensive loss function is established to describe the difference information between objects.Then,according to the Bayesian decision rules and the pre-given initial cost function,the calculation method of adaptive threshold pairs is constructed.An attribute reduction algorithm for minimizing decision cost in interval-valued decision making systems is established.Finally,an example is given,and the results show that the proposed reduction algorithm can obtain the reduction result with less decision cost,and is more rational in theory.

关 键 词：区间值决策系统 三支决策 代价函数 距离度量 属性约简 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]