基于正区域和差别元素的增量式属性约简算法  

Incremental Algorithm for Attribute Reduction Based on Positive Region and Discernibility Element

作　　者：刘涛涛[1] 马福民[1] 张腾飞[2] 

机构地区：[1]南京财经大学信息工程学院,南京210023 [2]南京邮电大学自动化学院,南京210023

出　　处：《计算机工程》2016年第8期183-187,193,共6页Computer Engineering

基　　金：国家自然科学基金资助项目(61403184;61105082);国家电子商务信息处理国际联合研究基金资助项目(2013B01035);江苏省自然科学基金资助项目(BK2012470);江苏省政府留学基金资助项目(JS-2013-342)

摘　　要：决策表数据的动态变化要求约简结果不断更新,为了在保证得到最简约简结果的同时提高属性约简的效率,提出一种改进的决策表约简算法,得到一个和原决策表等价的简化决策表。在此基础上,综合正区域和差别矩阵两种思想并利用原决策表的约简结果,给出一种仅存储由新增对象所产生的差别元素的增量式属性约简算法。实例计算结果表明,该算法能在原决策表约简结果的基础上快速更新属性约简结果。The reduction result should be updated continually with the dynamic changing of data in decision table. In order to improve the efficiency of attribute reduction while ensuring the simplest results, an improved decision table reduction algorithm is introduced to acquire a simplified decision table which is equivalent to the original one. Based on that, combining the merits of positive region and discernibility matrix, this paper proposes an incremental algorithm for attribute reduction, which makes full use of the reduction results of the original decision table and only stores the discernibility elements generated by the inserted new objects. Example calculation results show that the proposed algorithm can quickly update attribute reduction results on the basis of the reduction results of the original decision table.

关 键 词：属性约简 简化决策表 差别矩阵 正区域 增量式更新 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]