Thumbnail - No thumbnail

Sparse mixture of von Mises-Fisher distribution

Barbaro, Florian; Rossi, Fabrice (2021), Sparse mixture of von Mises-Fisher distribution, ESANN 2021 - European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, i6doc.com, p. 263-268. 10.14428/esann/2021.ES2021-115

Type
Communication / Conférence
External document link
https://www.esann.org/sites/default/files/proceedings/2021/ES2021-115.pdf
Date
2021
Conference title
29th European Symposium on Artificial Neutral Networks, Computational Intelligence and Machine Learning
Conference date
2021-10
Conference city
OnLine
Book title
ESANN 2021 - European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning
Publisher
i6doc.com
ISBN
978287587082-7
Pages
263-268
Publication identifier
Metadata
Show full item record
Author(s)
Barbaro, Florian
Statistique, Analyse et Modélisation Multidisciplinaire (SAmos-Marin Mersenne) [SAMM]
Rossi, Fabrice
CEntre de REcherches en MAthématiques de la DEcision [CEREMADE]
Abstract (EN)
Mixtures of von Mises-Fisher distributions can be used to cluster data on the unit hypersphere. This is particularly adapted for high-dimensional directional data such as texts. We propose in this article to estimate a von Mises mixture using a l1 penalized likelihood. This leads to sparse prototypes that improve both clustering quality and interpretability. We introduce an expectation-maximisation (EM) algorithm for this estimation and show the advantages of the approach on real data benchmark. We propose to explore the trade-off between the sparsity term and the likelihood one with a simple path following algorithm.