摘要
基于Rough Set理论中的不可分辨性原理,给出两个新的定义属性的最大区分值(Maximum Dis-cernibility Value,MDV)和属性冗余度(Attribute Redundancy Rate,ARR)。在数据预处理阶段,属性的MDV数值用于确定关于自组织映射网络SOM输出单元数量的启发式搜索策略;属性冗余度则用于衡量属性约简结果的信息冗余程度,并以此作为优化SOM网络输出层结构的依据。不依赖于领域经验知识,建立了MDV、SOM、ARR的组合算法模型,实现了Rough Set理论中连续属性的自动离散化计算,并明显提高了属性约简的速度。最后,通过项目实例对全过程进行有效验证。
In this paper,based on the indiscernibility discipline in Rough Set theory,two new measurement definitions are defined;attribute Maximum Discernibility Value （MDV） and Attribute Redundancy Rate （ARR）. MDV is introduced to decide the heuristic strategy for the Self-Organizing feature Map （SOM） neural network in the data preprocessing stage. And the attribute redundancy rate is for the attribute reduction as a effective feedback to the SOM clustering. Independent of domain experience,the combination of MDV,SOM, Skowron reduction,and the ARR can adjust the clustering number for every continuous attirbute automatically. Therefore ,in theory ,the computational speed is heightened greatly for the rough set attribute reduction. And in the end,a virtual project application is demostrated for the whole process effectively.
出处
《广西师范大学学报（自然科学版）》
CAS
北大核心
2006年第4期46-49,共4页
Journal of Guangxi Normal University:Natural Science Edition
基金
Sino-German Goverment Cooperation Project (2002DFG00027)
