J4 ›› 2007, Vol. 29 ›› Issue (2): 86-88.
• 论文 • 上一篇 下一篇
王天江 晏伟峰 漆志旺
出版日期:
发布日期:
Online:
Published:
摘要：
属性约简的目的在于减少条件属性中不必要属性的数目,是知识发现中的关键问题之一。本文提出了一种改进的基于Rough集的启发式算法（IMSA）,定义了新的启发函数（WSH）。这个函数考虑了所有隐藏规则集的质量,并且考虑了相关规则集的权重。在算法本身的时间复杂度没有增加的前提下,能够解决MSA算法遇到多个相同MSH值时无法处理的 情况。实验分析表明,该算法是有效的。
关键词: 粗糙集 属性约简 启发函数 核
Abstract:
The purpose of attribute reduction is to reduce the number of unnecessary attributes.Attribute reduction is the key problem in knowledge discovery.Thi s paper presents an improved heuristic algorithm based on rough sets for attribute reduction,which is called Improved Maximum Support Algorithm.We define a new heuristic function called Weight Support Heuristic.It considers the overall quality of the set of potential rules and the weight of related rule s.The IMSA overcomes the demerits of MSA when MSA falls across some of the same values of the MSH function.But,the time complexity does not increase.The experimental results show that this algorithm is effective in the attribute reduction of decision tables.
Key words: （rough set,attribute reduction,heuristic function,core）
王天江 晏伟峰 漆志旺. 一种基于Rough Set的启发式属性约简算法[J]. J4, 2007, 29(2): 86-88.
0 / / 推荐
导出引用管理器 EndNote|Ris|BibTeX
链接本文: http://joces.nudt.edu.cn/CN/
http://joces.nudt.edu.cn/CN/Y2007/V29/I2/86
