摘要
信息系统是不确定数据的重要模型,多源信息系统可用来表示来自多个同构异源的复杂数据。本文研究了多源信息系统中的不确定性度量,基于信息粒化思想给出了四种度量工具来度量多源信息系统的不确定性。首先,通过不同的信息值类型定义了两个信息值之间的距离,从而诱导出多源信息系统中的相容类。其次,基于粒计算的思想,将这些相容类看作多源信息系统的信息颗粒。从而,通过信息颗粒引入多源信息系统中的四种不确定性度量工具,并研究其之间的联系。这些研究结果对建立信息系统的粒计算框架以及属性约简将会很有帮助。
Information system is an important model of uncertain data,a multi-source information system can be used to represent complex data from multiple isomorphic and heterogeneous sources.This paper studies uncertainty measures in a multi-source information system,and gives four measurement tools that are based on information granulation.Firstly,distances between two information values are defined by different information value types,which can induce some tolerance classes.Next,these tolerance classes are seen as information granule of the multi-source information system from the point of view of granular computing.Finally,uncertainty measure tools are introduced through information granules,and the relationships between them are studied.These results will be very helpful for establishing granular computing framework and attribute reduction of information systems.
作者
黄丹
蓝家新
吴发乾
HUANG Dan;LAN Jia-xin;WU Fa-qian(School of Computer Science and Engineering,Yulin Normal University,Yulin 537000,China;School of Mathematics and Statistics,Baise University,Baise 533000,China;School of Education Sciences,Guangxi Science&Technology Normal University,Laibin 546100,China)
出处
《模糊系统与数学》
北大核心
2023年第2期165-174,共10页
Fuzzy Systems and Mathematics
基金
2020年广西高校中青年教师科研基础能力提升项目(2020KY19012,2020KY19014)
关键词
不确定性度量
多源信息系统
相容类
粒计算
信息颗粒
约简
Uncertainty Measure
Multi-source Information System
Tolerance Class
Granular Computing
Information Granule
Reduction
