This contribution proposes a model-based classifier developed for compositional data. A full mixture of experts model with Dirichlet components is used to incorporate information both on the composition and on a set of covariates. Estimation issues are dealt with by a Bayesian approach, allowing the researcher to use the posterior distribution of the parameters to measure the classification uncertainty.
Ascari, R., Migliorati, S. (2021). A full mixture of experts model to classify constrained data. In CLADAG 2021 BOOK OF ABSTRACTS AND SHORT PAPERS : 13th Scientific Meeting of the Classification and Data Analysis Group Firenze, September 9-11, 2021 (pp.247-250). Firenze Univ Press [10.36253/978-88-5518-340-6].
A full mixture of experts model to classify constrained data
Ascari,R
;Migliorati, S
2021
Abstract
This contribution proposes a model-based classifier developed for compositional data. A full mixture of experts model with Dirichlet components is used to incorporate information both on the composition and on a set of covariates. Estimation issues are dealt with by a Bayesian approach, allowing the researcher to use the posterior distribution of the parameters to measure the classification uncertainty.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
