Evaluation of topographic clustering and its kernelization

Abstract : We consider the topographic clustering task and focus on the problem of its evaluation, which enables to perform model selection: topographic clustering algorithms, from the original Self Organizing Map to its extension based on kernel (STMK), can be viewed in the unified framework of constrained clustering. Exploiting this point of view, we discuss existing quality measures and we propose a new criterion based on an F-measure, which combines a compacity with an organization criteria and extend it to their kernel-based version.
Document type :
Conference papers
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-01531082
Contributor : Lip6 Publications <>
Submitted on : Thursday, June 1, 2017 - 11:15:09 AM
Last modification on : Thursday, March 21, 2019 - 2:39:46 PM

Links full text

Identifiers

Citation

Marie-Jeanne Lesot, Florence d'Alché-Buc, Georges Siolas. Evaluation of topographic clustering and its kernelization. ECML03 - 14th European Conference on Machine Learning, Sep 2003, Dubrovnik, Croatia. pp.265-276, ⟨10.1007/978-3-540-39857-8_25⟩. ⟨hal-01531082⟩

Share

Metrics

Record views

107