首页 | 本学科首页   官方微博 | 高级检索  
     

信息表属性约简新方法
引用本文:张迎春,张丹枫,闫德勤. 信息表属性约简新方法[J]. 计算机工程, 2010, 36(1): 49-50,72
作者姓名:张迎春  张丹枫  闫德勤
作者单位:(1. 辽宁师范大学计算机与信息技术学院,大连 116029;2. 沈阳医学院护理学院,沈阳 110034)
基金项目:基金项目:国家自然科学基金资助项目(60372071);中国科学院自动化研究所复杂系统与智能科学重点实验室开放课题基金资助项目(20070101)
摘    要:在研究区分能力大小的基础上建立一个用于指导信息表的绝对属性约简的粗糙集模型,研究区分能力和分类能力之间的关系,提出决策依赖区分精度概念,为指导决策表的相对属性约简提供了一个新的判据。给出区分精度、近似精度和决策依赖区分精度在属性约简过程中相互关系的研究结论,通过一组对比实验说明决策依赖区分精度比近似精度对分类能力的描述更细致客观。

关 键 词:区分精度  决策依赖区分精度  近似精度
修稿时间: 

New Methods for Attribute Reduction of Information Table
ZHANG Ying-chun,ZHANG Dan-feng,YAN De-qin. New Methods for Attribute Reduction of Information Table[J]. Computer Engineering, 2010, 36(1): 49-50,72
Authors:ZHANG Ying-chun  ZHANG Dan-feng  YAN De-qin
Affiliation:1. School of Computer and Information Technology, Liaoning Normal University, Dalian ! 16029; 2. Schoo! of Nursing, Shenyang Medical College, Shenyang 110034)
Abstract:A rough set model is established to supervise the absolute attribute reduction for information table on the basis of studying the separating capacity. And a novel conception is proposed, which is called discernibility quality based on decision, on the basis of exploring the relations between the ability of discernibility and classifying, and it is an important criterion to supervise the relative attribute reduction for decision table. Several related conclusions are drawn by theoretical analyses in studying discernibility quality, approximate quality and discernibility quality based on decision. Comparison experiment shows that discernibility quality based on decision is finer than approximate quality for describing the ability of classifying.
Keywords:disceinibility quality  discernibility quality based on decision  approximate quality
本文献已被 维普 等数据库收录!
点击此处可从《计算机工程》浏览原始摘要信息
点击此处可从《计算机工程》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号