truecluster


Why truecluster?
Truecluster is a framework for robust scalable clustering with model selection. It allows objective determination of the optimal number of clusters even for non-parametric cluster algorithms. This problem has not been considered as solved for more than 50 years since R. L. Thorndike's funny publication: "Who belongs in the family?" (Psychometrika, 4:267-276, 1953).

Publications
Oehlschlägel, J. 2007. Truecluster: robust scalable clustering with model selection, available at arXiv.org.
Oehlschlägel, J. 2007. Truecluster matching, available at arXiv.org.

Examples
truecluster examples
example data

Contact
info@truecluster.com

Legal
Terms of Use
Privacy Policy
Impressum

Last changed 24.5.2018
© - Copyright 1998-2007 J. Oehlschlägel
truecluster logo
www.truecluster.com
Terms of Use
Privacy Policy
Impressum