Enhancing Biomedical Document Ranking with Domain Knowledge Incorporation in a Multi-Stage Retrieval Approach. - Recherche d’Information et Synthèse d’Information
Communication Dans Un Congrès Année : 2024

Enhancing Biomedical Document Ranking with Domain Knowledge Incorporation in a Multi-Stage Retrieval Approach.

Résumé

This article presents the results we obtained during BioASQ Task 12B Phase A on document ranking. Our strategy is based on a two-stage retrieval approach composed of a retriever and a reranker. The retriever is based on BM25 scoring and RM3 query expansion. The ranker is a BERT cross-encoder pre-trained on a biomedical corpus (BioLinkBERT). We study the impact of incorporating domain knowledge (MeSH) into this Pretrained Language Model and build a voting system to combine the insights from multiple models. Independently from the challenge, we also investigate a way to reduce the number of input tokens to bypass BERT limitation of 512 tokens for the input sequence.
Fichier principal
Vignette du fichier
paper-15.pdf (1020.85 Ko) Télécharger le fichier
Origine Fichiers éditeurs autorisés sur une archive ouverte
licence

Dates et versions

hal-04744454 , version 1 (21-10-2024)

Licence

Identifiants

  • HAL Id : hal-04744454 , version 1

Citer

Maël Lesavourey, Gilles Hubert. Enhancing Biomedical Document Ranking with Domain Knowledge Incorporation in a Multi-Stage Retrieval Approach.. 12th BioASQ Workshop at CLEF 2024, Sep 2024, Grenoble, France. ⟨hal-04744454⟩
0 Consultations
0 Téléchargements

Partager

More