Header logo is ei

From outliers to prototypes: Ordering data




We propose simple and fast methods based on nearest neighbors that order objects from high-dimensional data sets from typical points to untypical points. On the one hand, we show that these easy-to-compute orderings allow us to detect outliers (i.e. very untypical points) with a performance comparable to or better than other often much more sophisticated methods. On the other hand, we show how to use these orderings to detect prototypes (very typical points) which facilitate exploratory data analysis algorithms such as noisy nonlinear dimensionality reduction and clustering. Comprehensive experiments demonstrate the validity of our approach.

Author(s): Harmeling, S. and Dornhege, G. and Tax, D. and Meinecke, F. and Müller, K-R.
Journal: Neurocomputing
Volume: 69
Number (issue): 13-15
Pages: 1608-1618
Year: 2006
Month: August
Day: 0

Department(s): Empirical Inference
Bibtex Type: Article (article)

Digital: 0
DOI: 10.1016/j.neucom.2005.05.015
Language: en
Organization: Max-Planck-Gesellschaft
School: Biologische Kybernetik

Links: PDF


  title = {From outliers to prototypes: Ordering data},
  author = {Harmeling, S. and Dornhege, G. and Tax, D. and Meinecke, F. and M{\"u}ller, K-R.},
  journal = {Neurocomputing},
  volume = {69},
  number = {13-15},
  pages = {1608-1618},
  organization = {Max-Planck-Gesellschaft},
  school = {Biologische Kybernetik},
  month = aug,
  year = {2006},
  month_numeric = {8}