| 基于模拟退火思想的优化k-means算法 |
| |
| 引用本文： | 陈慧萍,贺会景,陈岚峰,蒋峰.基于模拟退火思想的优化k-means算法[J].河海大学常州分校学报,2006,20(4):29-33. |
| |
| 作者姓名： | 陈慧萍 贺会景 陈岚峰 蒋峰 |
| |
| 作者单位： | 河海大学,计算机及信息工程学院,江苏,常州,213022 |
| |
| 摘 要： | 鉴于典型的基于划分的聚类算法——k-means算法中存在局部最优和算法执行速度慢等问题,提出了基于模拟退火思想的优化k!means算法.该算法将模拟退火思想用于对k-means算法的优化,是一种具有全局最优解和较高执行效率的算法.针对聚类算法典型数据集和随机产生的数据集,在不同情况下进行对比实验.实验结果表明,优化k-means算法优于基本的k-means算法。
|
| 关 键 词： | 数据挖掘 聚类分析 k!means算法 模拟退火 |
| 文章编号： | 1009-1130(2006)04-0029-04 |
| 修稿时间： | 2006年7月3日 |
Optimized k-means Algorithm Based on Simulated Annealing |
| |
| CHEN Hui-ping,HE Hui-jing,CHEN Lan-feng,JIANG Feng.Optimized k-means Algorithm Based on Simulated Annealing[J].Journal of Hohai University Changzhou,2006,20(4):29-33. |
| |
| Authors: | CHEN Hui-ping HE Hui-jing CHEN Lan-feng JIANG Feng |
| |
| Abstract: | Because of the problems such as local optimization and low efficiency of typical clustering algorithm based on partitioning k-means,an optimized k-means algorithm based on simulated annealing idea is proposed in this paper.Simulated annealing idea is used to improve the k-means algorithm in order to make the algorithm overall optimization and high efficiency.Different tests on different conditions are given on some typical datasets which are used in clustering algorithms and on some randomly generated datasets.The results show that the optimized algorithm exceeds basic k-means algorithm in some aspects. |
| |
| Keywords: | data mining cluster analysis k-means algorithm simulated annealing |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
