Sciweavers

SDM
2004
SIAM

An Abstract Weighting Framework for Clustering Algorithms

13 years 5 months ago
An Abstract Weighting Framework for Clustering Algorithms
act Weighting Framework for Clustering Algorithms Richard Nock Frank Nielsen Recent works in unsupervised learning have emphasized the need to understand a new trend in algorithmic design, which is to influence the clustering via weights on the instance points. In this paper, we handle clustering as a constrained minimization of a Bregman divergence. Theoretical results show benefits resembling those of boosting algorithms, and bring new modified weighted versions of clustering algorithms such as k-means, expectation-maximization (EM) and k-harmonic means. Experiments display the quality of the results obtained, and corroborate the advantages that subtle data reweightings may bring to clustering.
Richard Nock, Frank Nielsen
Added 31 Oct 2010
Updated 31 Oct 2010
Type Conference
Year 2004
Where SDM
Authors Richard Nock, Frank Nielsen
Comments (0)