属性序下基于优势关系灰色信息系统约简-《计算机技术与发展》

文章信息/Info

Title:: Attributes Reduction of Gray Information System Based on Dominance Relation under Attributes Order

Author(s):: GUO Ning; LIN He; TAN Jing; [1]College of Information Science and Engineering,Lanzhou University[2]No.72959 Union of PLA

Keywords:: dominance relation; attribute order; gray information system; reduction

摘要:: 粗糙集和灰色理论在数据挖掘领域各有优点,它们最终目标都是为了发现知识。将粗糙集和灰色系统结合,研究了区间灰色信息系统在优势关系下的约简方法。在现实世界中,属性之间可能存在着一种属性序的关系,将这种属性序关系引入到基于优势关系的灰色信息系统中,给出了两种属性约简的算法：一种基于分辨矩阵的算法,另一种不基于分辨矩阵的算法。较好地适应了不完备信息系统和大数据集信息系统约简。最后通过实例对比,得出两种算法的约简结果是一致的,并分析了各自的适用范围

Abstract:: Rough sets and grey theory have advantages in the field of data mining,but their ultimate goal is to discover knowledge.Combined rough set theory and gray theory,and studied interval gray information system based on dominance relation.In the real world attributes order relationship may exist,so it introduced such attributes order to the dominance relation-based gray information systems.It gave two attribute reduction algorithms： one algorithm based on discernibility matrix,another algorithm not depend on discernibility matrix.The two algorithms adapted to the incomplete information system and large data sets information system for reduction well.Finally,by comparing the results of the two algorithms on an example,the two algorithms obtained consistent reduction results,and the application scopes of the two algorithms were analyzed