Sciweavers

Share
PKDD
2004
Springer

Shape and Size Regularization in Expectation Maximization and Fuzzy Clustering

9 years 6 months ago
Shape and Size Regularization in Expectation Maximization and Fuzzy Clustering
The more sophisticated fuzzy clustering algorithms, like the Gustafson–Kessel algorithm [11] and the fuzzy maximum likelihood estimation (FMLE) algorithm [10] offer the possibility of inducing clusters of ellipsoidal shape and different sizes. The same holds for the EM algorithm for a mixture of Gaussians. However, these additional degrees of freedom often reduce the robustness of the algorithm, thus sometimes rendering their application problematic. In this paper we suggest shape and size regularization methods that handle this problem effectively.
Christian Borgelt, Rudolf Kruse
Added 02 Jul 2010
Updated 02 Jul 2010
Type Conference
Year 2004
Where PKDD
Authors Christian Borgelt, Rudolf Kruse
Comments (0)
books