摘要
如何在不完备信息系统中利用已有的数据进行属性约简、获取属性权重一直是粗糙集理论中的一个研究重点。提出了一种在不完备信息系统中基于相容关系和PCA(主成份分析)相结合的粗糙集分析方法,这种方法属于启发式的数据分析方法。给出了这种分析方法的算法和一个算例。结果表明,该方法能够较准确地确定决策表的核,并用定量分析的方法获取属性权重。
How to acquire the optimal attributes reduction and weighs of attributes in incomplete information system is a hot topic for researching in rough sets theory recently. A rough set analysis method based on tolerance matrix and principle component analyze was provided in incomplete information system, which is one of heuristic data analyzing methods. Corresponding algorithm and an example were also given in this article. The method provided in this paper was compared with other methods. The result shows the method provided in this paper can find efficiently the core of an incomplete information system and acquire weighs of attributes through quantitative analyses.
出处
《计算机科学》
CSCD
北大核心
2009年第8期268-269,287,共3页
Computer Science
关键词
相容关系
主成份分析(PCA)
粗糙集
属性约简
Tolerance matrix, Principle component analyze, Rough sets, Attributes reduction
