On the Potential Inadequacy of Mutual Information for Feature Selection: Proceedings of the 20th International Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2012)

Benoît Frénay, Gauthier Doquire, Michel Verleysen

Résultats de recherche: Contribution dans un livre/un catalogue/un rapport/dans les actes d'une conférenceArticle dans les actes d'une conférence/un colloque

Résumé

Despite its popularity as a relevance criterion for feature selection, the mutual information can sometimes be inadequate for this task. Indeed, it is commonly accepted that a set of features maximising the mutual information with the target vector leads to a lower probability of misclassification. However, this assumption is in general not true. Justifications and illustrations of this fact are given in this paper.
langue originaleAnglais
titreProceedings of the 20th International Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2012)
Editeuri6doc.com
Etat de la publicationPublié - 2012
Modification externeOui

Empreinte digitale Examiner les sujets de recherche de « On the Potential Inadequacy of Mutual Information for Feature Selection: Proceedings of the 20th International Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2012) ». Ensemble, ils forment une empreinte digitale unique.

  • Contient cette citation

    Frénay, B., Doquire, G., & Verleysen, M. (2012). On the Potential Inadequacy of Mutual Information for Feature Selection: Proceedings of the 20th International Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2012). Dans Proceedings of the 20th International Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2012) i6doc.com.