| 基于相对距离的密度聚类算法* |
| |
| 引用本文： | 何孝金,傅彦,陈安龙.基于相对距离的密度聚类算法*[J].计算机应用研究,2009,26(4):1335-1337. |
| |
| 作者姓名： | 何孝金 傅彦 陈安龙 |
| |
| 作者单位： | 电子科技大学,计算机科学与工程学院,成都,610054 |
| |
| 基金项目： | 国家“863”计划资助项目（2006AA01Z414） |
| |
| 摘 要： | 首先介绍传统距离计算方法在聚类应用中的不足，并针对这点提出一种基于权重向量的相对距离计算方法。在应用DBSCAN算法的基础上，融入相对距离的计算及k-d树的范围查找的应用。该算法不仅能得到很好的聚类效果，而且消除了数据的度量单位对聚类结果的影响。
|
| 关 键 词： | 相对距离 DBSCAN算法 多维二进制搜索树 聚类 |
Density of clustering algorithm based on relative distance |
| |
| HE Xiao-jin,FU Yan,CHEN An-long.Density of clustering algorithm based on relative distance[J].Application Research of Computers,2009,26(4):1335-1337. |
| |
| Authors: | HE Xiao-jin FU Yan CHEN An-long |
| |
| Affiliation: | (School of Computer Science & Engineering, University of Electronic Science & Technology of China, Chengdu 610054, China) |
| |
| Abstract: | This paper firstly introduced the shortage of the traditional method in calculating the deficiencies in the application of clustering. To address this shortcoming, put forward a relative distance calculation method based on the weighted vector. The algorithm was based on the DBSCAN algorithm, and integrated of the calculation of relative distance and the application of the scope finding of the k-d tree. The algorithm can not only be good clustering effect, and the elimination of the unit of measurement data on the impact of cluster results. |
| |
| Keywords: | relative distance DBSCAN algorithm k-d tree clustering |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《计算机应用研究》浏览原始摘要信息 |
|
点击此处可从《计算机应用研究》下载全文 |
