|
Published Articles >> Table of Contents >> Abstract
Fourth IEEE International Conference on Data Mining (ICDM'04)
pp. 11-18
Subspace Selection for Clustering High-Dimensional Data
Christian Baumgartner, University for Health Sciences, Austria
Claudia Plant, University for Health Sciences, Austria
Karin Kailing, University of Munich, Germany
Hans-Peter Kriegel, University of Munich, Germany
Peer Kroger, University of Munich, Germany
Full Article Text:
  
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/ICDM.2004.10112
Send link to a friend
| Abstract |
|
In high-dimensional feature spaces traditional clustering algorithms tend to break down in terms of efficiency and quality. Nevertheless, the data sets often contain clusters which are hidden in various subspaces of the original feature space. In this paper, we present a feature selection technique called SURFING (SUbspaces Relevant For clusterING) that finds all subspaces interesting for clustering and sorts them by relevance. The sorting is based on a quality criterion for the interestingness of a subspace using the k-nearest neighbor distances of the objects. As our method is more or less parameterless, it addresses the unsupervised notion of the data mining task "clustering" in a best possible way. A broad evaluation based on synthetic and real-world data sets demonstrates that SURFING is suitable to find all relevant subspaces in high dimensional, sparse data sets and produces better results than comparative methods.
|
 Additional Information
|
Citation:
Christian Baumgartner, Claudia Plant, Karin Kailing, Hans-Peter Kriegel, Peer Kroger,
"Subspace Selection for Clustering High-Dimensional Data,"
icdm,
pp. 11-18,
Fourth IEEE International Conference on Data Mining (ICDM'04),
2004
|
|
