关键词: 半监督学习, 成对约束, 半监督谱聚类, 距离矩阵

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