基于矩阵的动态局部相容粗糙集的增量方法  

The incremental mechanisms of dynamic local tolerance rough sets based on matrix

作　　者：周悦丽 林国平 谢淋淋 Zhou Yueli;Lin Guoping;Xie Linlin(School of Mathematics and Statistics,Minnan Normal University,Zhangzhou,363000,China;Institute of Meteorological Big Data-Digital Fujian,Zhangzhou,363000,China;Key Laboratory of Grain Calculation in Fujian Province,Zhangzhou,363000,China)

机构地区：[1]闽南师范大学数学与统计学院,漳州363000 [2]数字福建气象大数据研究所,漳州363000 [3]福建省粒计算重点实验室,漳州363000

出　　处：《南京大学学报（自然科学版）》2022年第3期519-531,共13页Journal of Nanjing University（Natural Science）

基　　金：国家自然科学基金(11871259);福建省自然科学基金面上项目(2021J01983,2021J01979)。

摘　　要：粗糙集模型作为一种重要的粒计算模型,是处理数据的重要工具.在实际生活中,由于数据来源的多样性,信息系统常出现集值型数据,这些信息系统被称为集值信息系统.由于信息的更新,集值信息系统中的属性集会发生动态变化,因此,基于局部相容粗糙集模型,研究用矩阵来表示其上、下近似的方法,讨论随着属性集的动态变化局部关系矩阵的变化以及上、下近似的变化,并通过具体实例说明提出的更新方法在处理集值型数据时的有效性.最后给出与增量方法对应的算法,并在UCI数据库中选取了几组数据进行实验.实验结果证明,这种通过矩阵表示上、下近似并对其进行更新的方法是有效的,可以提高计算效率,降低时间复杂度.As for an important granular computing model,rough set model is an important tool to deal with data. Information systems often have set-valued data because of the diversity of data sources in real life. They are called set-valued information systems. Due to the updating of information,the attribute set in a set-valued information system changes dynamically.Therefore,based on the local tolerance rough sets,a matrix-based approach to computing the upper and lower approximations of local tolerance rough sets is constructed. At the same time,this paper discusses the variations of local relation matrices and the upper and lower approximations of local tolerance rough sets with the dynamic change of the attribute set. In addition,the concrete examples are given to verify the effectiveness of the proposed mechanisms in processing set-valued data. Finally,the algorithms corresponding to incremental mechanisms are given and experiments are carried out by using some data sets from UCI. Experimental results illustrate that the matrix representations of the upper and lower approximations are effective,which improve the calculation efficiency and reduce the time complexity while updating the upper and lower approximations.

关 键 词：相容关系 局部粗糙集 矩阵 集值信息系统 

分 类 号：O29[理学—应用数学] TP18[理学—数学]