Computer Engineering and Applications ›› 2014, Vol. 50 ›› Issue (15): 177-180.

Previous Articles     Next Articles

Research of constraints-expansion semi-supervised spectral clustering algorithm

SUN Guanghui, PAN Meisen   

  1. School of Computer, Hunan University of Arts and Science, Changde, Hunan 415000, China
  • Online:2014-08-01 Published:2014-08-04

扩展约束的半监督谱聚类算法研究

孙光辉,潘梅森   

  1. 湖南文理学院 计算机学院,湖南 常德 415000

Abstract: Based on several typical clustering algorithm analysis and comparison, this paper proposes a new clustering based on constraint expansion(CESSC). This algorithm expands the known constraints set, changes the similarity relation of the sample points through the density-sensitive path distance, and then combines with semi-supervised spectral clustering to cluster. Experimental results on UCI benchmark data sets prove that CESSC algorithm has good clustering effect.

Key words: semi-supervised learning, pair-wise constraint, semi-supervised spectral clustering, distance matrix

摘要: 通过对几种典型聚类算法的分析和比较,提出了一种新的聚类算法,基于扩展约束的半监督谱聚类算法,简称CE-SSC。这种算法扩展了已知约束集,通过密度敏感距离改变样本点的相似关系,结合半监督谱聚类进行聚类。在UCI基准集上的仿真实验结果证明,基于扩展约束的半监督谱聚类算法具有良好的聚类效应。

关键词: 半监督学习, 成对约束, 半监督谱聚类, 距离矩阵