冗余数据约简的研究与设计  

Study and Design of Reduction for Redundant Data

作　　者：纪霞[1] 李龙澍[2] 

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

出　　处：《计算机工程》2007年第5期62-64,共3页Computer Engineering

基　　金：国家自然科学基金资助项目(60273043);安徽省自然科学基金资助项目(050420204);安徽省高校选拔类人才基金资助项目

摘　　要：Z.Pawlak于1982年提出的Rough集理论有效地分析了不确定、不精确、不一致等各种不完备信息,其优点是无需任何关于数据的初始的或附加的信息,如统计学中的概率分布。该文介绍了Rough集的基本理论在数据约简中的应用。在分析基于信息系统的粗糙集理论的基础上,描述了一种基于核与重要度的约简算法,从降低约简算法计算复杂度角度出发,修改了属性约简算法,计算了算法修改前后的复杂度。实验结果表明,修改后的算法在降低时间复杂度的同时得出了次优属性集的约简。Advanced by Z. Pawlak in 1982, the rough set theory analyzes inadequate information efficiently such as inexactitude or dissonant information. The advantage of the rough set theory is no need of any initial or extra information such as the probability dispersal of statistics and so on. This article mainly introduces the applications of the rough set theory in data reduction. On the basis of analyzing the rough set theory based on information system, description of attribute set reduction algorithm based on the core and importance degree is given. In order to reduce the computational complexity of the algorithm, an improved algorithm is put forward. A brief comparison of computation complexity of the old algorithm with the improved one is made. This paper tests the improved algorithm with practical data. The result shows that the improved algorithm reduces the computational complexity and gains the solution inferior to the best attribute reduction in most cases.

关 键 词：粗集理论 约筒算法 重要度 

分 类 号：TP311.52[自动化与计算机技术—计算机软件与理论]