Volume 16, 2012
|Page(s)||436 - 452|
|Published online||04 September 2012|
How the initialization affects the stability of the қ-means algorithm∗
Centre de Recerca Matemàtica, Barcelona, Spain
2 University of Washington, Department of Statistics, Seattle, U.S.A.
3 Max Planck Institute for Biological Cybernetics, Tübingen, Germany
Revised: 2 May 2012
We investigate the role of the initialization for the stability of the қ-means clustering algorithm. As opposed to other papers, we consider the actual қ-means algorithm (also known as Lloyd algorithm). In particular we leverage on the property that this algorithm can get stuck in local optima of the қ-means objective function. We are interested in the actual clustering, not only in the costs of the solution. We analyze when different initializations lead to the same local optimum, and when they lead to different local optima. This enables us to prove that it is reasonable to select the number of clusters based on stability scores.
Mathematics Subject Classification: 62F12
Key words: Clustering / қ-means / stability / model selection
© EDP Sciences, SMAI, 2012
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.