• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2014, Vol. 36 ›› Issue (03): 524-529.

• 论文 • 上一篇    下一篇

基于约束投影的近邻传播聚类算法

钱雪忠1,赵建芳1,贾志伟2   

  1. (1.江南大学物联网工程学院,江苏 无锡 214122;2.成都信息工程学院,四川 成都 610225)
  • 收稿日期:2012-09-04 修回日期:2012-12-21 出版日期:2014-03-25 发布日期:2014-03-25
  • 基金资助:

    国家自然科学基金资助项目(61103129);江苏省科技支撑计划资助项目(BE2009009)

Constraint projection based affinity propagation         

QIAN Xuezhong 1,ZHAO Jianfang1,JIA Zhiwei2   

  1. (1.School of Internet of Things Engineering,Jiangnan University,Wuxi 214122;
    2.Chengdu University of Information Technology,Chengdu 610225,China)
  • Received:2012-09-04 Revised:2012-12-21 Online:2014-03-25 Published:2014-03-25

摘要:

提出了一种基于约束投影的近邻传播AP聚类算法。AP算法是在数据点相似度矩阵的基础上进行聚类的,很多传统的聚类方法都无法与其相媲美。但是,对于结构复杂的数据,AP算法往往得不到理想的结果。文中算法先对约束信息进行扩展,然后利用扩展的约束信息指导投影矩阵的获取,在低维空间中,利用约束信息对聚类结果进行修正。实验表明,文中算法与对比算法相比,时间性能更优,聚类效果更佳。

关键词: 半监督, 聚类, 约束信息, 投影, 近邻传播

Abstract:

A clustering algorithm, named constraint projection based affinity propagation (AP), is proposed. The AP algorithm conducts clustering based on similarity matrix, outperforming many traditional clustering algorithms. However, for those datasets with complex structure, the AP algorithm cannot always achieve the ideal results. Firstly, constraints are enlarged. Secondly, the enlarged constrains are used in getting the projection matrix. At last, the clustering result is updated by the enlarged constraints in the space with lower dimension. The result shows that, compared with the comparison algorithms, the proposal is better in both time performance and clustering results.

Key words: semi-supervised;clustering;constraints;projection;affinity propagation