TU Darmstadt / ULB / TUbiblio

Language classification from bilingual word embedding graphs

Eger, Steffen ; Hoenen, Armin ; Mehler, Alexander (2016)
Language classification from bilingual word embedding graphs.
26th International Conference on Computational Linguistics (COLING 2016). Osaka, Japan (11.12.2016-16.12.2016)
Konferenzveröffentlichung, Bibliographie

Kurzbeschreibung (Abstract)

We study the role of the second language in bilingual word embeddings in monolingual semantic evaluation tasks. We find strongly and weakly positive correlations between down-stream task performance and second language similarity to the target language. Additionally, we show how bilingual word embeddings can be employed for the task of semantic language classification and that joint semantic spaces vary in meaningful ways across second languages. Our results support the hypothesis that semantic language similarity is influenced by both structural similarity as well as geography/contact.

Typ des Eintrags: Konferenzveröffentlichung
Erschienen: 2016
Autor(en): Eger, Steffen ; Hoenen, Armin ; Mehler, Alexander
Art des Eintrags: Bibliographie
Titel: Language classification from bilingual word embedding graphs
Sprache: Englisch
Publikationsjahr: Dezember 2016
Ort: Osaka, Japan
Verlag: The COLING 2016 Organizing Committee
Buchtitel: Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: Technical Papers
Veranstaltungstitel: 26th International Conference on Computational Linguistics (COLING 2016)
Veranstaltungsort: Osaka, Japan
Veranstaltungsdatum: 11.12.2016-16.12.2016
URL / URN: http://www.aclweb.org/anthology/C16-1331
Kurzbeschreibung (Abstract):

We study the role of the second language in bilingual word embeddings in monolingual semantic evaluation tasks. We find strongly and weakly positive correlations between down-stream task performance and second language similarity to the target language. Additionally, we show how bilingual word embeddings can be employed for the task of semantic language classification and that joint semantic spaces vary in meaningful ways across second languages. Our results support the hypothesis that semantic language similarity is influenced by both structural similarity as well as geography/contact.

Freie Schlagworte: UKP_reviewed;UKP_a_DLinNLP
ID-Nummer: TUD-CS-2016-1446
Fachbereich(e)/-gebiet(e): 20 Fachbereich Informatik
20 Fachbereich Informatik > Ubiquitäre Wissensverarbeitung
Hinterlegungsdatum: 31 Dez 2016 14:29
Letzte Änderung: 02 Jul 2024 09:36
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