基于仿射传播聚类的K-means算法优化
首发时间:2015-06-03
摘要:传统的K-means算法的初始聚类中心是随机选择的,其聚类结果随不同的初始聚类中心选择而波动。针对K-means算法对初始聚类中心敏感的问题,本文提出了基于仿射传播算法改进的K-means算法,该算法利用仿射传播算法来确定初始聚类中心,然后通过K-means进行聚类。实验结果表明,改进的算法比原算法能取得更好的聚类效果。
关键词: 仿射传播聚类 K-means 聚类中心
For information in English, please click here
Optimization of K-means Algorithm Based on Affinity Propagation Algorithm
Abstract:The initial clustering center of K-means algorithm is chosen at random. With different clustering center input, the clustering results vary. The K-means algorithm is sensitive to the initial clustering center problem. This paper proposes an improved K-means algorithm based on affinity propagation algorithm. This algorithm uses affine propagation algorithm to determine the initial cluster center, then cluster by K-means. The experimental results show that, the improved algorithm can get better clustering results.
Keywords: affinity propagation algorithm K-means initial clustering center
基金:
论文图表:
引用
No.4644922106534914****
同行评议
共计0人参与
勘误表
基于仿射传播聚类的K-means算法优化
评论
全部评论0/1000