| 基于关联规则的特征选择方法 |
| |
| 引用本文： | 何小晨,徐守时.基于关联规则的特征选择方法[J].红外与激光工程,2002,31(6):504-507. |
| |
| 作者姓名： | 何小晨 徐守时 |
| |
| 作者单位： | 中国科学技术大学电子工程与信息科学系,安徽,合肥,230027 |
| |
| 摘 要： | 提出了一种基于关联规则的特征选择方法，在市场分析中，人们用关联规则来发掘大量数据集中项目之间的关系。关联规则的方法非常适用于图像分析中的特征选择，可以自动地选出可信度较高的特征。使用这一方法挑选合适的纹理特征用于图像分类，并对结果进行分析。
|
| 关 键 词： | 特征选择 关联规则 纹理特征 监督分类 数字图像处理 |
| 文章编号： | 1007-2276(2002)06-0504-04 |
| 收稿时间： | 2002/3/15 |
| 修稿时间： | 2002年3月15日 |
Method of feature selection based on association rules |
| |
| HE Xiao\|chen,XU Shou\|shi.Method of feature selection based on association rules[J].Infrared and Laser Engineering,2002,31(6):504-507. |
| |
| Authors: | HE Xiao\|chen XU Shou\|shi |
| |
| Abstract: | A new method of feature selection based on association rules is proposed. Association rules have been used in application such as market analysis to capture relationships among items in large data sets. It is shown that association rules can be adapted to feature selection which are used in image analysis and they can automatically identify the most confident feature. Appropriate texture features are selected with this method to serve image classification and simulative results are discussed. |
| |
| Keywords: | Feature selection Association rules Texture feature Supervisory classification |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
