Improving stacking methodology for combining classifiers: applications to cosmetic industry

Charles Gomes Hisham Nocairi Marie Thomas Gilbert Saporta 1
1 CEDRIC - MSDMA - CEDRIC. Méthodes statistiques de data-mining et apprentissage
CEDRIC - Centre d'études et de recherche en informatique et communications
Abstract : 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.
Document type :
Journal articles
Complete list of metadatas

Cited literature [34 references]  Display  Hide  Download
Contributor : Gilbert Saporta <>
Submitted on : Sunday, February 9, 2020 - 11:21:07 AM
Last modification on : Thursday, February 13, 2020 - 1:26:44 AM


Publisher files allowed on an open archive




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⟩



Record views


Files downloads