Is document frequency important for PRF?

Abstract : We introduce in this paper a new heuristic constraint for PRF models, referred to as the Document Frequency (DF) constraint, which is validated through a series of experiments with an oracle. We then analyze, from a theoretical point of view, state-of-the-art PRF models according to their relation with this constraint. This analysis reveals that the standard mixture model for PRF in the language modeling family does not satisfy the DF constraint on the contrary to several recently proposed models. Lastly, we perform tests, which further validate the constraint, with a simple family of tf-idf functions based on a parameter controlling the satisfaction of the DF constraint.
Type de document :
Communication dans un congrès
Giambattista Amati and Fabio Crestani. ICTIR 2011 - International Conference on the Theory Information Retrieval, Sep 2011, Bertinoro, Italy. Springer-Verlag, pp.89-100, 2011
Liste complète des métadonnées

Littérature citée [15 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-00742242
Contributeur : Eric Gaussier <>
Soumis le : mardi 16 octobre 2012 - 11:46:41
Dernière modification le : jeudi 11 janvier 2018 - 06:27:15
Document(s) archivé(s) le : jeudi 17 janvier 2013 - 11:35:16

Fichier

Clinchant-ictir2011.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-00742242, version 1

Collections

Citation

Stéphane Clinchant, Éric Gaussier. Is document frequency important for PRF?. Giambattista Amati and Fabio Crestani. ICTIR 2011 - International Conference on the Theory Information Retrieval, Sep 2011, Bertinoro, Italy. Springer-Verlag, pp.89-100, 2011. 〈hal-00742242〉

Partager

Métriques

Consultations de la notice

134

Téléchargements de fichiers

82