× Description Téléchargement Publication(s) Contact
 Retourner à Logiciels et Ressources

ELMoLex

Analyseur syntaxique neuronal développé pour la soumission d'ALMAnaCH à la shared task CoNLL-18 d'analyse syntaxique multilingue

Site web principal

Description

Neural parsing system developed for ALMAnaCH's submission to the CoNLL-18 multilingual parsing shared task. ELMoLex uses Deep Biaffine Parser as its backbone and extends it with UDLexicons-based lexical features.

Citation et publication(s)

Si vous utilisez ce travail, merci de citer :

Ganesh Jawahar, Benjamin Muller, Amal Fethi, Louis Martin, Éric Villemonte de La Clergerie, Benoît Sagot and Djamé Seddah. 2018. ELMoLex: Connecting ELMo and Lexicon features for Dependency Parsing.
In CoNLL 2018 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies. Brussels, Belgium.
HAL PDF
@inproceedings{jawahar:hal-01959045,
 address = {Brussels, Belgium},
 author = {Jawahar, Ganesh and Muller, Benjamin and Fethi, Amal and Martin, Louis and Villemonte de La Clergerie, {\'E}ric and Sagot, Beno{\^i}t and Seddah, Djam{\'e}},
 title = {{ELMoLex: Connecting ELMo and Lexicon features for Dependency Parsing}},
year = {2018},
 booktitle = {{CoNLL 2018 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies}},
 doi = {10.18653/v1/K18-2023},
 url = {https://inria.hal.science/hal-01959045},
 hal_pdf = {https://inria.hal.science/hal-01959045v1/file/elmolex-connecting-elmo.pdf},
}

Contact

Pour plus d'informations ou pour poser une question, merci de contacter Benjamin Muller et Djamé Seddah

benjamin.muller[at]inria.fr et djame.seddah[at]inria.fr