User menu

Accès à distance ? S'identifier sur le proxy UCLouvain

Feature Scoring by Mutual Information for Classification of Mass Spectra

Primary tabs

Krier, Catherine [UCL] François, Damien [UCL] Wertz, Vincent [UCL] Verleysen, Michel [UCL]

Selecting relevant features in mass spectra analysis is important both for classification and search for causality. In this paper, it is shown how using mutual information can help answering to both objectives, in a model-free nonlinear way. A combination of ranking and forward selection makes it possible to select several feature groups that may lead to similar classification performances, but that may lead to different results when evaluated from an interpretability perspective.

Document type Communication à un colloque (Conference Paper) – Présentation orale avec comité de sélection
Access type Accès restreint
Publication date 2006
Language Anglais
Conference "7th International FLINS Conference on Applied Artificial Intelligence (FLINS 2006)", Genova (Italy) (du 29/08/2006 au 31/08/2006)
Peer reviewed yes
Host document "Proceedings of the 7th International FLINS Conference on Applied Artificial Intelligence (FLINS 2006)"- 557-564
Collection World Scientific
Publisher D. Ruan, P. D'Hondt, P.F. Fantoni, M. De Cock, M. Nechtegael, E.E. Kerre eds
Affiliations UCL - FSA/ELEC - Département d'électricité
UCL - FSA/INMA - Département d'ingénierie mathématique
Links
Bibliographic reference Krier, Catherine ; François, Damien ; Wertz, Vincent ; Verleysen, Michel. Feature Scoring by Mutual Information for Classification of Mass Spectra.7th International FLINS Conference on Applied Artificial Intelligence (FLINS 2006) (Genova (Italy), du 29/08/2006 au 31/08/2006). In: Proceedings of the 7th International FLINS Conference on Applied Artificial Intelligence (FLINS 2006), D. Ruan, P. D'Hondt, P.F. Fantoni, M. De Cock, M. Nechtegael, E.E. Kerre eds2006, p.557-564
Permanent URL http://hdl.handle.net/2078.1/90752