Pregled bibliografske jedinice broj: 1274167
Measuring Terminology Consistency in Translated Corpora: Implementation of the Herfindahl-Hirshman Index
Measuring Terminology Consistency in Translated Corpora: Implementation of the Herfindahl-Hirshman Index // Information, 13 (2022), 2; 43-57 doi:10.3390/info13020043 (međunarodna recenzija, članak, znanstveni)
CROSBI ID: 1274167 Za ispravke kontaktirajte CROSBI podršku putem web obrasca
Naslov
Measuring Terminology Consistency in Translated
Corpora: Implementation of the Herfindahl-Hirshman
Index
Autori
Gašpar, Angelina ; Seljan, Sanja ; Kučiš, Vlasta
Izvornik
Information (2078-2489) 13
(2022), 2;
43-57
Vrsta, podvrsta i kategorija rada
Radovi u časopisima, članak, znanstveni
Ključne riječi
terminology consistency ; parallel corpora ; translation ; legislation ; information transfer ; Herfindahl-Hirshman Index (HHI)
Sažetak
Consistent terminology can positively influence communication, information transfer, and proper understanding. In multilingual written communication processes, challenges are augmented due to translation variants. The main aim of this study was to implement the Herfindahl-Hirshman Index (HHI) for the assessment of translated terminology in parallel corpora for the evaluation of translated terminology. This research was conducted on three types of legal domain subcorpora, dating from different periods: the Croatian-English parallel corpus (1991–2009), Latin-English and Latin-Croatian versions of the Code of Canon Law (1983), and English and Croatian versions of the EU legislation (2013). After the terminology extraction process, validation of term candidates was performed, followed by an evaluation. Terminology consistency was measured using the HHI—a commonly accepted measurement of market concentration. Results show that the HHI can be used for measuring terminology consistency to improve information transfer and message understanding. In translation settings, the process shows the need for quality management solutions.
Izvorni jezik
Engleski
Znanstvena područja
Informacijske i komunikacijske znanosti
POVEZANOST RADA
Projekti:
--11-931-1072 - Strojno učenje i obrada prirodnog jezika u domeni računalne sigurnosti (Seljan, Sanja) ( CroRIS)
Ustanove:
Filozofski fakultet, Zagreb
Citiraj ovu publikaciju:
Časopis indeksira:
- Web of Science Core Collection (WoSCC)
- Emerging Sources Citation Index (ESCI)
- Scopus