计算机工程 ›› 2008, Vol. 34 ›› Issue (14): 60-62.doi: 10.3969/j.issn.1000-3428.2008.14.022

• 软件技术与数据库 • 上一篇    下一篇

基于均衡化函数的k均值优化算法

钱雪忠1,施培蓓1,张明阳2,汪 中3   

  1. (1. 江南大学信息工程学院,无锡 214122;2. 华东计算技术研究所,上海 200233;3. 中国科技大学计算机科学与技术系,合肥 230027)

  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2008-07-20 发布日期:2008-07-20

Optimized k-means Algorithm Based on Balanced Function

QIAN Xue-zhong1, SHI Pei-bei1, ZHANG Ming-yang2, WANG Zhong3   

  1. (1. School of Information Engineering, Southern Yangtze University, Wuxi 214122;2. East-China Institute of Computer Technology, Shanghai 200233;3. Department of Computer Science, University of Science and Technology of China, Hefei 230027)

  • Received:1900-01-01 Revised:1900-01-01 Online:2008-07-20 Published:2008-07-20

摘要: 传统的k-means算法要求用户事先给定k值,限制了很多应用,初始中心点随机选择,容易导致局部极值点,常用的评价函数对于求解最优聚类数目也不理想。针对这些问题,该文提出一种新的评价函数——均衡化函数,采用基于密度的初始化中心点选择算法,自动生成聚类数目,实验结果表明了改进算法的有效性。

关键词: k-均值算法, 密度, 初始中心点, 均衡化函数

Abstract: In traditional k-means algorithm, value k must be confirmed in advance, which restricts a large number of practical applications. Initial centers are selected randomly so that local extremums will be introduced. The common evaluate functions to the optimum number of clustering can not be effectively calculated. To conquer these problems, a new evaluation function-equalization function based on the density of the center initialization algorithms introduced, and the number of generation clustering are automatically calculated. Experimental results prove the efficiency of the improved k-means algorithm.

Key words: k-means algorithm, density, initial center point, balanced function

中图分类号: