|
|||||
|
|
| 基于Relief算法的特征学习聚类 | |
| 引用本文： | 吴艳文,胡学钢,陈效军.基于Relief算法的特征学习聚类[J].合肥学院学报(自然科学版),2008,18(2):45-48. |
| 作者姓名： | 吴艳文 胡学钢 陈效军 |
| 作者单位： | 1. 巢湖职业技术学院,安徽,巢湖,238000 2. 合肥工业大学,计算机与信息学院,合肥,230009 3. 合肥学院,基础教学部,合肥,230022 |
| 摘 要： | 聚类作为数据挖掘常用工具之一，是按照事物间的相似性进行的一种无监督分类．然而传统的聚类方法较少考虑特征权值．为此，通过研究、分析Relief算法及其在聚类应用中存在的问题，提出了一种基于Relief算法的特征评价函数，并将此函数运用到特征学习聚类中，以解决特征权值取值不当对聚类产生的负面影响． |
| 关 键 词： | 特征评价函数 Relief算法 特征学习聚类 |
| 文章编号： | 1673-162X(2008)02-0045-04 |
| 修稿时间： | 2007年11月15 |
Feature Learning Clustering Based on Relief Algorithm |
|
| WU Yan-wen,Hu Xue-gang,Chen Xiao-jun.Feature Learning Clustering Based on Relief Algorithm[J].Journal of Hefei University :Natural Sciences,2008,18(2):45-48. | |
| Authors: | WU Yan-wen Hu Xue-gang Chen Xiao-jun |
| Institution: | WU Yan-wen, Hu Xue-gang,Chen Xiao-jun( 1. Chaohu Vocational and Technical collage, Chaohu, Anhui 238000;2. School of Computer and Information, Hefei Uni versity of Technology, Hefei 230009;3. Department of Basic Courses, Hefei University, Hefei 230022, China) |
| Abstract: | Clustering unsupervised classification according to similarity of objects is one of common tools in data mining. Unfortunately, many traditional clustering algorithms assumed the distribution of each feature is uniformed. By researching and analyzing relief algorithms and its weakness in clustering, a novel feature criterion function based on relief algorithm has been proposed in the paper. Meanwhile, we apply the function into feature learning clustering in order to counteract the negative affects by the given feature weighting wrongly. |
| Keywords: | feature criterion function relief algorithm feature learning clustering |
| 本文献已被 维普 万方数据 等数据库收录！ | |