基于非对称相似差别矩阵知识约简  

Knowledge reduction based on asymmetrical similar matrix

作　　者：王加阳[1] 杜库[1] 胡沛[1] 高灿[2] 

机构地区：[1]中南大学信息科学与工程学院,长沙410083 [2]同济大学计算机科学与技术,上海200092

出　　处：《计算机工程与应用》2010年第7期44-45,82,共3页Computer Engineering and Applications

基　　金：湖南省科技计划项目No.2008FJ3184;湖南省自然科学基金No.06JJ20075~~

摘　　要：信息系统中存在着大量数据值缺省的情况,为寻求约简的最优解需耗费大量的时间。用非对称相似关系代替粗糙集理论中的等价关系,定义了非对称相似差别矩阵,提出了基于非对称相似差别矩阵的高效求核和知识约简算法。该算法无需改变初始不完备信息系统的结构,能直接处理缺省数据。实验结果表明,新算法所获得的决策规则简洁、高效,与缺省值无关。In many kinds of information systems,there exists a situation where a value of data is null and algorithms given in many papers to calculate its optimal reduction have a high space complexity which is a big problem.To solve this problem,the equivalence relation of rough set theory is replaced by an asymmetrical similar relation,and from this relation,an asymmetrical similar matrix is defined,then a new heuristic algorithm with less space complexity that require core and knowledge reduction based on asymmetrical similar matrix is proposed,and an example is given to confirm the validity of the algorithm.

关 键 词：粗糙集 非对称相似关系 差别矩阵 知识约简 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]