Evaluating Automatic Term Extraction Methods on Individual Documents (CROSBI ID 684616)
Prilog sa skupa u zborniku | izvorni znanstveni rad | međunarodna recenzija
Podaci o odgovornosti
Šajatović, Antonio ; Buljan, Maja ; Šnajder, Jan ; Dalbelo Bašić, Bojana
engleski
Evaluating Automatic Term Extraction Methods on Individual Documents
Automatic Term Extraction (ATE) extracts terminology from domain-specific corpora. ATE is used in many NLP tasks, including Computer Assisted Translation, where it is typically applied to individual documents rather than the entire corpus. While corpus-level ATE has been extensively evaluated, it is not obvious how the results transfer to documentlevel ATE. To fill this gap, we evaluate 16 state-of-the-art ATE methods on full-length documents from three different domains, on both corpus and document levels. Unlike existing studies, our evaluation is more realistic as we take into account all gold terms. We show that no single method is best in corpuslevel ATE, but C-Value and KeyConceptRelatendess surpass others in document-level ATE.
Automatic term extraction ; Computer Assisted Translation ; Evaluation
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
Podaci o prilogu
149-154.
2019.
objavljeno
10.18653/v1/W19-5118
Podaci o matičnoj publikaciji
Proceedings of the Joint Workshop on Multiword Expressions and WordNet (MWE-WN 2019)
Firenza : München:
Podaci o skupu
7th Workshop on Balto-Slavic Natural Language Processing. Association for Computational Linguistics
poster
28.07.2019-02.08.2019
Firenca, Italija