Rough set feature extraction by remarkable degrees with real world decision-making problems
Rough set feature extraction by remarkable degrees with real world decision-making problems
复制标题
DOI：
10.1007/s00500-009-0494-1
复制
发表时间：
2010-10
期刊：
影响因子：
4.1
通讯作者：
P. Guo
中科院分区：
文献类型：
--
作者：
P. Guo
本文提出了一种利用粗糙集方法对决策表中的IF-THEN规则进行简化的新方法。该方法能够显著地提取不同决策类对象的主要特征。利用该方法对两个实际决策问题进行了研究。一是分析日本产业的主要特征。二是对日本老年人的生活状况进行剖析。分析结果表明，该方法能有效地从属性和对象较多的决策表中提取决策类的主要特征。获得的主要特征可以深入了解对象的情况，并有助于在现实世界中做出决策。
In this paper, a novel approach for simplifying the obtained if–then rules from decision table by rough set based methods is proposed. This approach can extract main features of objects in different decision classes by remarkable degrees. Using the proposed method, two real world decision-making problems are studied. The first one is for analyzing the main features of Japanese industries. The second is for anatomizing the life situations of senior citizens in Japan. The analysis results show that the proposed method is powerful for piecing out the main features of decision classes in the decision table which has many attributes and objects. The obtained main features can provide deep insight into the situations of objects and are useful for making a decision in the real world.