Recherche de groupes parallèles en classification non-supervisée

Abstract : In this paper we focus on an unsupervised classification case, where clusters share a common "shape". We consider that this shape consists of a given hyperplane, common to all clusters up to a given a translation. Points are thus considered as distributed around a set of parallel hyperplanes. The underlying objective function can be seen as minimizing the sum of distances of each point to its hyperplane. Similarly to k-means, this goal is achieved by alternating affectation- (of each point to an hyperplane) and computation- (of the hyperplane equation) phases. Seeking for parallel hyperplanes, this computation phase is conducted simultaneously for all hyperplanes.
Complete list of metadatas

Cited literature [10 references]  Display  Hide  Download
Contributor : Matthieu Exbrayat <>
Submitted on : Monday, January 4, 2016 - 2:00:29 PM
Last modification on : Thursday, February 7, 2019 - 5:02:18 PM
Long-term archiving on : Friday, April 15, 2016 - 4:03:35 PM


Files produced by the author(s)


  • HAL Id : hal-01250182, version 1



Lionel Martin, Matthieu Exbrayat, Teddy Debroutelle, Aladine Chetouani, Sylvie Treuillet, et al.. Recherche de groupes parallèles en classification non-supervisée. 16ème Conférence Internationale Francphone sur l'Extraction et la Gestion des Connaissances EGC 2016, Jan 2016, Reims, France. ⟨hal-01250182⟩



Record views


Files downloads