Combining Language Models with NLP and Interactive Query Expansion. - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2010

Combining Language Models with NLP and Interactive Query Expansion.

Résumé

Following our previous participation in INEX 2008 Ad-hoc track, we continue to address both standard and focused retrieval tasks based on comprehensible language models and interactive query expansion (IQE). Query topics are expanded using an initial set of Multiword Terms (MWTs) selected from top n ranked documents. In this experiment, we extract MWTs from article titles, narrative field and automatically generated summaries. We combined the initial set of MWTs obtained in an IQE process with automatic query expansion (AQE) using language models and smoothing mechanism. We chose as baseline the Indri IR engine based on the language model using Dirichlet smoothing. We also compare the performance of bag of word approaches (TFIDF and BM25) to search strategies elaborated using language model and query expansion (QE). The experiment is carried out on all INEX 2009 Ad-hoc tasks.
Fichier principal
Vignette du fichier
inex2009_LNCS-final.pdf (125.58 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-00635286 , version 1 (25-10-2011)

Identifiants

Citer

Eric Sanjuan, Fidelia Ibekwe-Sanjuan. Combining Language Models with NLP and Interactive Query Expansion.. 8th International Workshop of the Initiative for the Evaluation of XML Retrieval (INEX 2009), Dec 2009, Brisbane, Australia. pp.122-132, ⟨10.1007/978-3-642-14556-8_14⟩. ⟨hal-00635286⟩
168 Consultations
539 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More