| 一种基于粗糙集的K-Means聚类算法 |
| |
| 引用本文： | 冯征.一种基于粗糙集的K-Means聚类算法[J].计算机工程与应用,2006,42(20):141-142,146. |
| |
| 作者姓名： | 冯征 |
| |
| 作者单位： | 天津大学管理学院,天津,300072 |
| |
| 摘 要： | 在传统的硬聚类过程中,得到的簇中数据对象是确定的,然而在现实世界,边界数据是不能被准确划分到任何一个簇的。粗糙集是处理这种边界不确定性的工具,基于此提出了一种基于粗糙集的K-Means聚类算法,这种算法生成的簇包括上近似集和下近似集,可以处理边界对象。试验证明,这种算法是有效的。
|
| 关 键 词： | 聚类 粗糙集 上近似集 下近似集 K均值 |
| 文章编号： | 1002-8331-（2006）20-0141-02 |
| 收稿时间： | 2005-11 |
| 修稿时间： | 2005-11 |
A Rough-based K-Means Clustering Algorithm |
| |
| Feng Zheng.A Rough-based K-Means Clustering Algorithm[J].Computer Engineering and Applications,2006,42(20):141-142,146. |
| |
| Authors: | Feng Zheng |
| |
| Affiliation: | School of Management,Tianjin University,Tianjin 300072 |
| |
| Abstract: | Traditional hard clustering get crisp boundary of each cluster,however,in reality the boundary of data is hard to partition.Rough set is an effective tool to deal with the uncertain boundary region.Based on it,the paper puts forward a rough-based K-means algorithm which gets the clusters including lower approximations subsets and upper approximations subsets to deal with boundary region.Data experiment shows the effectiveness of the algorithm. |
| |
| Keywords: | clustering rough set lower approximations upper approximations K-means |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
