| 基于密度和对象方向聚类算法的改进 |
| |
| 引用本文： | 孟海东,张玉英.基于密度和对象方向聚类算法的改进[J].计算机工程与应用,2006,42(20):154-156. |
| |
| 作者姓名： | 孟海东 张玉英 |
| |
| 作者单位： | 内蒙古科技大学网络中心,包头014010 |
| |
| 基金项目： | 内蒙古自治区高等教育科学研究项目（编号：NJ04019） |
| |
| 摘 要： | 针对K-means算法所存在的问题进行了深入的研究,提出了基于密度和聚类对象方向的改进算法(KADD算法)。该算法采取聚类对象分布密度方法来确定初始聚类中心,然后根据对象的聚类方向来发现任意形状的簇。理论分析与实验结果表明,改进算法在不改变时间、空间复杂度的情况下能取得更好的聚类结果。
|
| 关 键 词： | 数据挖掘 聚类 K-means算法 KADD算法 |
| 文章编号： | 1002-8331-（2006）20-0154-03 |
| 收稿时间： | 2005-10 |
| 修稿时间： | 2005-10 |
Improved Clustering Algorithm Meng Haidong Based on Density and Direction |
| |
| Zhang Yuying.Improved Clustering Algorithm Meng Haidong Based on Density and Direction[J].Computer Engineering and Applications,2006,42(20):154-156. |
| |
| Authors: | Zhang Yuying |
| |
| Affiliation: | Network Center, Inner Mongolia University of Science and Technology, Baotou 014010 |
| |
| Abstract: | The existing problems of K-means clustering algorithm are carefully researched.An improved K-means algorithm based on density and direction(KADD) is presented,with which initial clustering center points are located according to the clustering objects distribution density.And the clusters with arbitrary distributions are bind based on object direction.Theory analysis and experimental results demonstrate that the improved algorithm can get better clustering results without changing efficiency and dimensional complexity. |
| |
| Keywords: | data mining clustering K-means algorithm KADD algorithm |
| 本文献已被 CNKI 维普 等数据库收录！ |
