Outlier and novelty detection for Functional data: a semiparametric Bayesian approach

Francesco Denti, Andrea Cappozzo, Francesca Greselin

Risultato della ricerca: Contributo in libroContributo a convegno

Abstract

Given a sample of unlabeled observations, the goal of a novelty detec- tion method is to identify which units substantially deviate from the observed la- beled patterns. Therefore, in a model-based framework, it is firstly of paramount importance to learn the components that correspond to the manifest groups in the training set. Secondly, one needs to take into account the lack of knowledge regard- ing the statistical novelties. Thirdly, contaminated elements in the known classes could greatly jeopardize the identification of new groups. Motivated by these chal- lenges, we propose a two-stage Bayesian non-parametric novelty detector. At stage one, robust estimates are extracted from the training set and, subsequently, such in- formation is employed to elicit informative priors within a flexible semiparametric mixture. This general paradigm can be easily adapted to complex modeling frame- works: we provide here an application to functional data from a food authenticity study.
Lingua originaleEnglish
Titolo della pubblicazione ospiteBook of Short Papers of the 5th international workshop on Models and Learning for Clustering and Classification
Pagine33-38
Numero di pagine6
DOI
Stato di pubblicazionePubblicato - 2021
EventoMBC2 2020 - Catania
Durata: 31 ago 20202 set 2020

Convegno

ConvegnoMBC2 2020
CittàCatania
Periodo31/8/202/9/20

Keywords

  • Bayesian mixture model
  • Dirichlet Process Mixture Model
  • Functional data
  • Minimum Regularized Covariance Determinant

Fingerprint

Entra nei temi di ricerca di 'Outlier and novelty detection for Functional data: a semiparametric Bayesian approach'. Insieme formano una fingerprint unica.

Cita questo