 |
    |
Constrained Clustering with Local Constraint Propagation
Ping He, Xiaohua Xu, and Ling Chen
Department of Computer Science, College of Information Engineering, Yangzhou University, Yangzhou 225000, Chinaangeletx@springer.com
arterx@springer.com
yzulchen@springer.com
Abstract. We consider the problem of multi-class constrained clustering given pairwise constraints, which specify the pairs of data belonging to the same or different clusters. In this paper, we present a new constrained clustering algorithm, Local Constraint Propagation (LCP), which can propagate the influence of each pairwise constraint to the unconstrained data with sufficient smoothness. It not only reveals the underlying structures of the clusters, but also integrates the influence of all the pairwise constraints on every data point. Promising experiments on image segmentations demonstrate the effectiveness of our method.
Keywords: constrained clustering, constraint propagation, image segmentation
LNCS 7585, p. 223 ff.
lncs@springer.com
© Springer-Verlag Berlin Heidelberg 2012