Improving stacking methodology for combining classifiers: applications to cosmetic industry - Cnam - Conservatoire national des arts et métiers Accéder directement au contenu
Article Dans Une Revue Electronic Journal of Applied Statistical Analysis Année : 2016

Improving stacking methodology for combining classifiers: applications to cosmetic industry

Hisham Nocairi
  • Fonction : Auteur
  • PersonId : 964687
Marie Thomas
  • Fonction : Auteur
  • PersonId : 964688

Résumé

Stacking (Wolpert (1992), Breiman (1996)) is known to be a successful way of linearly combining several models. We modify the usual stacking methodology when the response is binary and predictions highly correlated,by combining predictions with PLS-Discriminant Analysis instead of ordinary least squares. For small data sets we develop a strategy based on repeated split samples in order to select relevant variables and ensure the robustness of the nal model. Five base (or level-0) classiers are combined in order to get an improved rule which is applied to a classical benchmark of UCI Machine Learning Repository. Our methodology is then applied to the prediction of dangerousness of 165 chemicals used in the cosmetic industry, described by 35 in vitro and in silico characteristics, since faced to safety constraints, one cannot rely on a single prediction method, especially when the sample sizeis low.
Fichier principal
Vignette du fichier
stackingEJASA.pdf (929.75 Ko) Télécharger le fichier
Origine : Fichiers éditeurs autorisés sur une archive ouverte
Loading...

Dates et versions

hal-02471754 , version 1 (09-02-2020)

Licence

Paternité - Pas d'utilisation commerciale - Pas de modification

Identifiants

Citer

Charles Gomes, Hisham Nocairi, Marie Thomas, Gilbert Saporta. Improving stacking methodology for combining classifiers: applications to cosmetic industry. Electronic Journal of Applied Statistical Analysis, 2016, 09 (2), pp.340 - 361. ⟨10.1285/i20705948v9n2p340⟩. ⟨hal-02471754⟩
166 Consultations
737 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More