Apprentissage de Modèles de Langue Neuronaux pour la Recherche d'Information

Nicolas Despres 1 Sylvain Lamprier 1 Benjamin Piwowarski 2
1 MLIA - Machine Learning and Information Access
LIP6 - Laboratoire d'Informatique de Paris 6
2 BD - Bases de Données
LIP6 - Laboratoire d'Informatique de Paris 6
Abstract : Information Retrieval (IR) faces different difficulties,notablythoserelatedtovocab- ulary mismatch issues and term dependencies. In the last few years, language models based on neural networks have been proposed to deal with both term dependencies and vocabulary mismatch issues in complex natural language processing tasks. However, to be efficient, these models require huge amounts of training data. They have thus never been employed for IR ad- hoc tasks directly, where the estimation of one language model per document is required. We propose an approach based on the specialization of a generic language model, learned on the whole document collection, by a set of document-specific parameters, to define neural language models fitted for ad-hoc IR tasks.
Document type :
Conference papers
Complete list of metadatas

https://hal.sorbonne-universite.fr/hal-01358680
Contributor : Benjamin Piwowarski <>
Submitted on : Thursday, September 1, 2016 - 11:25:36 AM
Last modification on : Thursday, March 21, 2019 - 1:21:22 PM

Identifiers

  • HAL Id : hal-01358680, version 1

Citation

Nicolas Despres, Sylvain Lamprier, Benjamin Piwowarski. Apprentissage de Modèles de Langue Neuronaux pour la Recherche d'Information. Conférence en Recherche d'Infomations et Applications, Mar 2016, Toulouse, France. pp.717-732. ⟨hal-01358680⟩

Share

Metrics

Record views

267