HAL will be down for maintenance from Friday, June 10 at 4pm through Monday, June 13 at 9am. More information
Skip to Main content Skip to Navigation
Conference papers

Résolution d'anaphores nominales avec les séparateurs à vastes marges sur les arbres syntaxiques

Abstract : Coreference resolution is the task of finding all expressions that refer to the same entity in a text. It is an important step for a lot of NLP applications that involve natural language under-standing. However syntactic knowledge is important for noun phrase resolution and is traditionally represented in terms of features vector selected and defined heuristically. In this paper we proposea model based on support vectors machine and kernel methods, applied to syntax trees (easier to construct and richer in information needed to resolve nominal anaphors). In this way, we avoid the effort involved in decoding these trees into feature vectors. The model obtained after training has been tested on a subset of the semEval task 1 corpus. The F-measures on the different evaluation metrics for the coreference resolution systems are respectively 48.36% for the MUC, 49.53% for the BLANC, 86.99% for the BCUB and 78.96% for the CEAF.
Document type :
Conference papers
Complete list of metadata

Cited literature [19 references]  Display  Hide  Download

Contributor : Dimedrik Feudjieu Connect in order to contact the contributor
Submitted on : Tuesday, September 1, 2020 - 10:57:45 AM
Last modification on : Tuesday, December 7, 2021 - 5:50:03 PM
Long-term archiving on: : Wednesday, December 2, 2020 - 1:21:11 PM


Files produced by the author(s)


  • HAL Id : hal-02926896, version 1


Dimedrik Feudjieu, Paulin Melatagia Yonta. Résolution d'anaphores nominales avec les séparateurs à vastes marges sur les arbres syntaxiques. CARI 2020 - Colloque Africain sur la Recherche en Informatique et en Mathématiques Appliquées, Oct 2020, Thiès, Sénégal. ⟨hal-02926896⟩



Record views


Files downloads