基于模糊粗糙集的辨识矩阵属性约简方法  

Fuzzy rough attribute reduction method based on dicernibility matrix

作　　者：赵晋欢 王长忠[1] ZHAO Jinhuan;WANG Changzhong(College of Mathematics and Physics,Bohai University,Jinzhou 121013,China)

机构地区：[1]渤海大学数理学院

出　　处：《渤海大学学报（自然科学版）》2019年第2期146-151,共6页Journal of Bohai University:Natural Science Edition

基　　金：国家自然科学基金项目(No:61572082);辽宁省自然科学基金项目(No:20170540012)

摘　　要：通过构造辨识矩阵进行属性约简是一种有效的降低数据维度的方法.然而,经典粗糙集构造的辨识矩阵的局限性在于并不适用于连续型数据,只适用于离散型数据.因此,本文研究在模糊粗糙集下的辨识矩阵属性约简模型,并讨论该模型的相关数学性质,提出相应的算法来对连续型数据进行属性约简.通过实例验证该方法是有效的.Constructing dicernibility matrix to attribute reduction is an effective method to reduce data dimension.However,the limitation of dicernibility matrix constructed by classical rough set is that it is not suitable for continuous data,but only for discrete data.Therefore,in this paper,we study fuzzy rough attribute reduction model based on dicernibility matrix,and discuss the mathematical properties of the model,so an algorithm for attribute reduction of continuous data is proposed.The method is validated by an example.

关 键 词：辨识矩阵 属性约简 模糊粗糙集 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]