Naslov
On Machine Translation of User Reviews

Autori
Popović, Maja ; Poncelas, Alberto ; Brkić Bakarić, Marija ; Way, Andy

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), ostalo

Izvornik
Proceedings of Recent Advances in Natural Language Processing ({; ; {; ; RANLP}; ; }; ; ) / Angelova, Galia ; Kunilovskaya, Maria ; Mitkov, Ruslan ; Nikolova-Koleva, Ivelina - , 2021, 1113-1122

ISBN
978-954-452-072-4

Skup
International Conference Recent Advances in Natural Language Processing (RANLP 2021)

Mjesto i datum
Online, 01.09.2021. - 03.09.2021

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
neural machine translation, user reviews genre, morphologically complex languages, out-of-domain parallel corpora, synthetic parallel corpora, evaluation

Sažetak
This work investigates neural machine translation (NMT) systems for translating English user reviews into Croatian and Serbian, two similar morphologically complex languages. Two types of reviews are used for testing the systems: IMDb movie reviews and Amazon product reviews. Two types of training data are explored: large out-of-domain bilingual parallel corpora, as well as small synthetic in-domain parallel corpus obtained by machine translation of monolingual English Amazon reviews into the target languages. Both automatic scores and human evaluation show that using the synthetic in-domain corpus together with a selected subset of out-of-domain data is the best option. Separated results on IMDb and Amazon reviews indicate that MT systems perform differently on different review types so that user reviews generally should not be considered as a homogeneous genre. Nevertheless, more detailed research on larger amount of different reviews covering different domains/topics is needed to fully understand these differences.

Izvorni jezik
Engleski

Znanstvena područja
Informacijske i komunikacijske znanosti

POVEZANOST RADA