The Community for Technology Leaders
Green Image
Issue No. 06 - June (2010 vol. 22)
ISSN: 1041-4347
pp: 900-905
Jaewook Lee , Pohang University of Science and Technology, Pohang
Daewon Lee , University of Ulsan, Ulsan
ABSTRACT
Clustering methods utilizing support estimates of a data distribution have recently attracted much attention because of their ability to generate cluster boundaries of arbitrary shape and to deal with outliers efficiently. In this paper, we propose a novel dissimilarity measure based on a dynamical system associated with support estimating functions. Theoretical foundations of the proposed measure are developed and applied to construct a clustering method that can effectively partition the whole data space. Simulation results demonstrate that clustering based on the proposed dissimilarity measure is robust to the choice of kernel parameters and able to control the number of clusters efficiently.
INDEX TERMS
Clustering, kernel methods, dynamical systems, equilibrium vector, support.
CITATION
Jaewook Lee, Daewon Lee, "Dynamic Dissimilarity Measure for Support-Based Clustering", IEEE Transactions on Knowledge & Data Engineering, vol. 22, no. , pp. 900-905, June 2010, doi:10.1109/TKDE.2009.140
108 ms
(Ver )