| 基于Rough集的连续属性离散化方法 |
| |
| 引用本文： | 尹旭日.基于Rough集的连续属性离散化方法[J].计算机工程与设计,2006,27(11):2038-2040. |
| |
| 作者姓名： | 尹旭日 |
| |
| 作者单位： | 解放军汽车管理学院,基础部,安徽,蚌埠,233011 |
| |
| 摘 要： | 连续属性离散化是Rough集理论应用中面临的主要问题之一.提出了一种基于的Rough集连续属性离散化方法.首先提出主泛化决策等概念,在数据过滤方法的基础上,利用等价类的合并对属性离散化.实验表明,利用该方法对数据进行离散预处理后提取的规则具有较好的分类预测准确性.
|
| 关 键 词： | 离散化 Rough集 泛化决策 等价类 置信度 |
| 文章编号： | 1000-7024（2006）11-2038-03 |
| 收稿时间： | 2005-04-08 |
| 修稿时间： | 2005-04-08 |
Discrete method of continuous attributes based on rough set |
| |
| YIN Xu-ri.Discrete method of continuous attributes based on rough set[J].Computer Engineering and Design,2006,27(11):2038-2040. |
| |
| Authors: | YIN Xu-ri |
| |
| Abstract: | The discretization of continuous attributes is one of the most main problems in applying of rough set theory.A new method of discretization based on rough set theory is presented.The concept of generalized decision is introduced.Its basic idea is to merge the equivalence classes of(Q on basis of the methods of data filtering.Experimental results show this algorithm is efficient,comprehensible and suitable for a pre-processing step before a data-mining method is employed. |
| |
| Keywords: | discretization Rough set generalized decision equivalence classes confidence degree |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
