区间值信息系统中一种极大相容类的增量更新算法  

Incremental Algorithm for Updating Maximal Consistent Classes in Interval-valued Information System

作　　者：张迎迎[1,2] 李天瑞[1,2] 

机构地区：[1]西南交通大学信息科学与技术学院,成都611756 [2]四川省云计算与智能技术重点实验室,成都611756

出　　处：《小型微型计算机系统》2017年第7期1573-1579,共7页Journal of Chinese Computer Systems

基　　金：国家自然科学基金项目(61573292)资助

摘　　要：由于区间值信息系统中对象的属性值都是区间值,经典粗糙集中的等价关系已不再适用.给出一个新的距离度量来定义属性值为区间值的两对象之间的距离,基于此给出了邻域关系及邻域的定义,然后给出了区间值信息系统中的极大相容类的定义.由于实际应用中信息系统是动态变化的,即新对象的加入和已有对象的删除,因此如何利用已求得的结果增量的更新极大相容类是亟待解决的一个关键问题.由此,提出一种基于Bron-Kerbosch算法的增量更新极大相容类的算法,实验结果表明,该算法能有效提高更新极大相容类的效率.Because the attribute values of objects in Interval-valued Information System （ IvlS ） are interval numbers, the equivalence relation in the classical rough set is not applicable in IvIS any more. In this paper, a new distance measure is given to define the distance between any two objects in IvIS. Based on this, the definitions of the neighborhood relation and the neighborhood are given, and then the definition of the maximal consistent class is presented. However, in practice, the information system changes dynamically. For example, some new objects are added and some existing objects are deleted. Therefore, how to utilize the obtained results to incrementally update maximal consistent classes is a key problem. So an incremental algorithm based on Bron-Kerbosch algorithm is proposed, and the experimental results show that the proposed algorithm can improve the efficiency of updating the maximal consistent classes.

关 键 词：区间值信息系统 粗糙集 极大相容类 Bron—Kerbosch算法 增量学习 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]