基于非平衡数据下不完备混合型信息系统的属性约简  

Attribute reduction of incomplete hybrid information system based on unbalanced data

作　　者：姚晟[1,2] 李初宴 陈悦 Yao Sheng;Li Chuyan;Chen Yue(School of Computer Science&Technology,Anhui University,Hefei 230039,China;Key Laboratory of Intelligent Computing&Signal Processing of Ministry of Education,Anhui University,Hefei 230039,China)

机构地区：[1]安徽大学计算机科学与技术学院,合肥230039 [2]安徽大学计算智能与信号处理教育部重点实验室,合肥230039

出　　处：《计算机应用研究》2021年第5期1331-1335,共5页Application Research of Computers

基　　金：国家自然科学基金资助项目(61602004,61300057);安徽省自然科学基金资助项目(1508085MF127);安徽省高等学校自然科学研究重点资助项目(KJ2016A041);安徽大学信息保障技术协同创新中心公开招标课题(ADXXBZ20145,ADXXBZ20146);安徽大学博士科研启动基金资助项目(J10113190072)。

摘　　要：完备混合型信息系统下的粗糙集模型是传统粗糙集模型的重要扩展,目前关于非平衡数据属性约简的研究仅限于完备混合型的粗糙集模型。针对这一问题,提出一种基于不完备混合型信息系统的非平衡数据属性约简。本文首先将传统的粗糙集模型进行推广,提出不完备混合型信息系统下的粗糙集模型;然后针对数据的非平衡性,根据上下边界区域和类分布的不均匀性定义了一种新的属性重要度;在基于区别矩阵的基础上设计出一种非平衡数据的属性约简算法。实验分析表明该算法针对不完备非平衡数据的属性约简具有一定的有效性和优越性。The rough set model under complete hybrid information system is an important extension of the traditional rough set model.However,the research on the attribute reduction of unbalanced data is limited to the complete hybrid rough set model.To solve this problem,this paper proposed an unbalanced data attribute reduction based on incomplete hybrid information system.This paper generalized the traditional rough set model,and then proposed the rough set model under incomplete hybrid information system.Then,it defined a new attribute importance according to the heterogeneity of upper and lower boundary regions and class distribution for the unbalanced data.Finally,this paper designed an algorithm for attribute reduction of unba-lanced data based on discernible matrix.The experimental results show that the algorithm is effective and advantageous for attribute reduction of incomplete and unbalanced data.

关 键 词：粗糙集 不完备混合型信息系统 非平衡数据 属性约简 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]