TY - GEN
T1 - A rival penalized EM algorithm towards maximizing weighted likelihood for density mixture clustering with automatic model selection
AU - CHEUNG, Yiu Ming
N1 - Copyright:
Copyright 2011 Elsevier B.V., All rights reserved.
PY - 2004
Y1 - 2004
N2 - How to determine the number of clusters is an intractable problem in clustering analysis. In this paper, we propose a new learning paradigm named Maximum Weighted Likelihood (MwL), in which the weights are designoble. Accordingly, we develop a novel Rival Penalized Expectation-Maximization (RPEM) algorithm, whose intrinsic rival penalization mechanism enables the redundant densities in the mixture to be gradually faded out during the learning. Hence, the RPEM can automatically select an appropriate number of densities in density mixture clustering. The experiments have shown the promising results.
AB - How to determine the number of clusters is an intractable problem in clustering analysis. In this paper, we propose a new learning paradigm named Maximum Weighted Likelihood (MwL), in which the weights are designoble. Accordingly, we develop a novel Rival Penalized Expectation-Maximization (RPEM) algorithm, whose intrinsic rival penalization mechanism enables the redundant densities in the mixture to be gradually faded out during the learning. Hence, the RPEM can automatically select an appropriate number of densities in density mixture clustering. The experiments have shown the promising results.
UR - http://www.scopus.com/inward/record.url?scp=10044261725&partnerID=8YFLogxK
U2 - 10.1109/ICPR.2004.1333852
DO - 10.1109/ICPR.2004.1333852
M3 - Conference proceeding
AN - SCOPUS:10044261725
SN - 0769521282
T3 - Proceedings - International Conference on Pattern Recognition
SP - 633
EP - 636
BT - Proceedings of the 17th International Conference on Pattern Recognition, ICPR 2004
A2 - Kittler, J.
A2 - Petrou, M.
A2 - Nixon, M.
T2 - Proceedings of the 17th International Conference on Pattern Recognition, ICPR 2004
Y2 - 23 August 2004 through 26 August 2004
ER -