| 三角不等式原理对聚类算法的改进 |
| |
| 引用本文： | 何春霞,常晋义.三角不等式原理对聚类算法的改进[J].常熟理工学院学报,2007,21(2):100-103. |
| |
| 作者姓名： | 何春霞 常晋义 |
| |
| 作者单位： | 常熟理工学院,计算机科学与工程系,江苏,常熟,215500 |
| |
| 摘 要： | 聚类分析是数据挖掘中的一个重要研究领域,面对大规模的、高维的数据,如何建立有效的聚类算法是目前一个研究热点。现已有多种直接和快速的聚类算法,但是当处理海量数据时,时间效率仍然有待提高。本文应用三角不等式原理,分别对TTSAS算法和k-means算法提出改进,避免其中冗余的距离计算,提高原算法效率。
|
| 关 键 词： | 三角不等式原理 聚类 TTSAS k-means |
| 文章编号： | 1008-2794（2007）02-0100-04 |
| 收稿时间： | 2006-12-27 |
| 修稿时间： | 2006年12月27日 |
Using the Triangle Inequality to Accelerate Cluster Algorithm |
| |
| HE Chun-xia,CHANG Jin-yi.Using the Triangle Inequality to Accelerate Cluster Algorithm[J].Journal of Changshu Institute of Technology,2007,21(2):100-103. |
| |
| Authors: | HE Chun-xia CHANG Jin-yi |
| |
| Abstract: | Cluster analysis is an important research field in the data mining,and it is a research focusing on how to set up effective cluster algorithm in the face of the large-scale,high-dimensional datasets.There are many kinds of very direct and fast cluster algorithms,but when the large numbers of data are handled,the efficiency still needs improving.This paper presents the improvements on the TTSAS algorithm and the k-means algorithm respectively.The new algorithms avoid unnecessary distance calculations by applying the triangle inequality.They are more effective for datasets and at the same time exactly the same results as the standard algorithms can still always be obtained. |
| |
| Keywords: | triangle inequality cluster TTSAS k-means |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
