摘要
决策粗糙集是目前粗糙集理论的重要研究分支。目前的决策粗糙集很少对不完备混合型的信息系统进行研究,为了改善这一局限,提出一种扩展的决策粗糙集模型。通过引入邻域容差关系来处理不完备混合型信息系统,在其基础上定义扩展的决策粗糙集模型,同时提出相应的三支决策。在该模型的基础上设计一种最小化决策代价的属性约简算法。根据三支决策,构建出一种不完备混合型数据的三支决策分类算法。实验结果表明,该算法具有更高的数据分类精度和更小的误分类代价。
Decision-theoretic rough set is an important branch of rough set theory.However,at present,decision-theoretic rough sets rarely study incomplete mixed information systems.In order to improve this limitation,this paper presents an extended decision-theoretic rough set model.The incomplete mixed information system was dealt with by introducing the neighborhood tolerance relation;an extended decision-theoretic rough set model was defined based on it,and the corresponding three-way decisions were also proposed;based on the proposed model,an attribute reduction algorithm was designed to minimize the decision cost;according to the proposed three-way decisions,a three-way decisions classification algorithm with incomplete mixed data was constructed.The simulation results show that the proposed classification algorithm has higher data classification accuracy and lower misclassification cost.
作者
王光琼
Wang Guangqiong(School of Intelligent Manufacturing,Sichuan University of Arts and Science,Dazhou 635000,Sichuan,China)
出处
《计算机应用与软件》
北大核心
2020年第11期246-254,共9页
Computer Applications and Software
基金
“2019年中国物流学会/中国物流与采购联合会面上研究课题计划”面上项目(2019CSLKT3-231)
四川省教育厅重点项目(18ZA0421)。
关键词
粗糙集
决策粗糙集
不完备混合型数据
三支决策
属性约简
分类
Rough set
Decision-theoretic rough set
Incomplete mixed data
Three-way decisions
Attribute reduction
Classification
