TU Darmstadt / ULB / TUbiblio

Language Transfer Learning for Supervised Lexical Substitution

Hintz, Gerold ; Biemann, Chris (2016)
Language Transfer Learning for Supervised Lexical Substitution.
Konferenzveröffentlichung, Bibliographie

Kurzbeschreibung (Abstract)

We propose a framework for lexical substitution that is able to perform transfer learning across languages. Datasets for this task are available in at least three languages (English, Italian, and German). Previous work has addressed each of these tasks in isolation. In contrast, we regard the union of three shared tasks as a combined multilingual dataset. We show that a supervised system can be trained effectively, even if training and evaluation data are from different languages. Successful transfer learning between languages suggests that the learned model is in fact independent of the underlying language. We combine state-of-the-art unsupervised features obtained from syntactic word embeddings and distributional thesauri in a supervised delexicalized ranking system. Our system improves over state of the art in the full lexical substitution task in all three languages.

Typ des Eintrags: Konferenzveröffentlichung
Erschienen: 2016
Autor(en): Hintz, Gerold ; Biemann, Chris
Art des Eintrags: Bibliographie
Titel: Language Transfer Learning for Supervised Lexical Substitution
Sprache: Deutsch
Publikationsjahr: August 2016
Buchtitel: Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics
URL / URN: http://www.aclweb.org/anthology/P16-1012
Zugehörige Links:
Kurzbeschreibung (Abstract):

We propose a framework for lexical substitution that is able to perform transfer learning across languages. Datasets for this task are available in at least three languages (English, Italian, and German). Previous work has addressed each of these tasks in isolation. In contrast, we regard the union of three shared tasks as a combined multilingual dataset. We show that a supervised system can be trained effectively, even if training and evaluation data are from different languages. Successful transfer learning between languages suggests that the learned model is in fact independent of the underlying language. We combine state-of-the-art unsupervised features obtained from syntactic word embeddings and distributional thesauri in a supervised delexicalized ranking system. Our system improves over state of the art in the full lexical substitution task in all three languages.

Freie Schlagworte: AIPHES_area_b3
ID-Nummer: TUD-CS-2016-0169
Fachbereich(e)/-gebiet(e): 20 Fachbereich Informatik
20 Fachbereich Informatik > Sprachtechnologie
DFG-Graduiertenkollegs
DFG-Graduiertenkollegs > Graduiertenkolleg 1994 Adaptive Informationsaufbereitung aus heterogenen Quellen
Hinterlegungsdatum: 30 Dez 2016 17:45
Letzte Änderung: 28 Sep 2018 15:33
PPN:
Export:
Suche nach Titel in: TUfind oder in Google
Frage zum Eintrag Frage zum Eintrag

Optionen (nur für Redakteure)
Redaktionelle Details anzeigen Redaktionelle Details anzeigen