摘要
属性约简是粗糙集的一个核心研究课题,但经典属性约简及其延伸算法是基于有决策属性的决策表的属性约简算法,它们对无决策属性的非常规决策表的属性约简无能为力。以粗糙集理论为基础,对无决策属性的非常规决策表从分形维数方面进行研究,提出了一种适用于无决策属性的决策表的启发式属性约简算法。该算法在一定程度上能够解决非常规决策表的属性约简问题,进一步扩展了粗糙集理论的应用范围。实例表明该算法是有效可行的。
Attribute reduction is a core research topic of rough set,but classical attribute reduction algorithm and its extend- ed algorithms are based on decision tables with decision attributes and can not be applied to attribute reduction of abnormal decision tables without decision attributes.Based on rough set theory,the abnormal decision tables in fractal dimension is stud- ied and a heuristic attribute reduction algorithm is presented.To a certain extent,the algorithm can resolve the attribute reduc- tion problem of abnormal decision tables and extend application of rough set theory.The example shows that the algorithm is effective and feasible
出处
《计算机工程与应用》
CSCD
北大核心
2011年第14期115-117,共3页
Computer Engineering and Applications
基金
河南省基础与前沿技术研究计划项目(No.102300410266)
郑州轻工业学院博士科研基金资助项目
关键词
属性约简
决策属性
决策表
分形维数
attribute reduction
decision attribute
decision table
fractal dimension
