Prédiction de la polysémie pour un terme biomédical

Juan Antonio Lossio-Ventura 1, 2, * Clement Jonquet 2 Mathieu Roche 1, 3 Maguelonne Teisseire 1, 3
* Corresponding author
1 ADVANSE - ADVanced Analytics for data SciencE
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier
2 SMILE - Système Multi-agent, Interaction, Langage, Evolution
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier
Abstract : Polysemy is the capacity for a term to have multiple meanings. Polysemy prediction is a first step for Word Sense Induction (WSI), which allows to find different meanings for a term, as well as for Information Extraction (IE) systems. In addition, the polysemy detection is important for building and enriching terminologies and ontologies. In this paper, we present a novel approach to detect if a biomedical term is polysemic or not, with the long term goal of enriching biomedical ontologies after disambiguation of candidate terms. This approach is based on meta-learning techniques, more precisely on meta-features. We propose the definition of novel meta-features, extracted directly from the text dataset, as well as from a graph of coc- current terms. Our method obtains very good results, with an Accuracy and F-mesure of 0.978.
Complete list of metadatas

Cited literature [14 references]  Display  Hide  Download

https://hal-lirmm.ccsd.cnrs.fr/lirmm-01132654
Contributor : Juan Antonio Lossio Ventura <>
Submitted on : Tuesday, March 17, 2015 - 4:23:06 PM
Last modification on : Wednesday, September 18, 2019 - 4:04:04 PM

File

CORIA2015.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : lirmm-01132654, version 1

Citation

Juan Antonio Lossio-Ventura, Clement Jonquet, Mathieu Roche, Maguelonne Teisseire. Prédiction de la polysémie pour un terme biomédical. CORIA: Conférence en Recherche d’Information et Applications, Mar 2015, Paris, France. pp.437-452. ⟨lirmm-01132654⟩

Share

Metrics

Record views

555

Files downloads

670