| 基于属性加权的改进K-Means算法 |
| |
| 引用本文： | 陈东,皮德常.基于属性加权的改进K-Means算法[J].数字社区&智能家居,2009(9). |
| |
| 作者姓名： | 陈东 皮德常 |
| |
| 作者单位： | 南京航空航天大学信息科学与技术学院;南京工业大学信息科学与工程学院; |
| |
| 摘 要： | 经典的K-Means算法认为被分析样本的各个属性对聚类结果的贡献均匀,没有考虑不同属性特征对聚类结果可能造成的不同影响。文章提出了一种基于样本属性加权的K-Means算法。该算法利用变异系数赋权法对属性进行加权处理,通过权值反映各个属性对聚类结果的贡献的大小。实验表明,该算法在不改变时间、空间复杂度的情况下能取得更好的聚类结果。
|
| 关 键 词： | 聚类 属性加权 K-Means |
Improved K-Means Algorithm based on the Attributes Weighted |
| |
| CHEN Dong ,PI De-chang.Improved K-Means Algorithm based on the Attributes Weighted[J].Digital Community & Smart Home,2009(9). |
| |
| Authors: | CHEN Dong PI De-chang |
| |
| Affiliation: | 1. College of information science and technology;Nanjing University of Aeronautics and Astronautics;Nanjing 210016;China;2. College of information science and engineering;Nanjing University of Technology;Nanjing 210009;China |
| |
| Abstract: | The classical K-Mmeans algorithm regards the attributes of swatches have the same contribute to the clustering result, and it does not think over the different attribute may have the different impact to the clustering result. In this paper, an improved K-Mmeans algorithm based on the attributes weighted is presented. It gives a different weight to every attribute using the variation coefficient method. The results of experiment demonstrate that the improved algorithm can get better clustering results withou... |
| |
| Keywords: | Clustering Attributes Weighted K-Means |
| 本文献已被 CNKI 等数据库收录！ |
