Lexical diversity in statistical and neural machine translation

Brglez, Mojca (avtor), Vintar, Špela (avtor)

Neural machine translation systems have revolutionized translation processes in terms of quantity and speed in recent years, and they have even been claimed to achieve human parity. However, the quality of their output has also raised serious doubts and concerns, such as loss in lexical variation, evidence of “machine translationese”, and its effect on post-editing, which results in “post-editese”. In this study, we analyze the outputs of three English to Slovenian machine translation systems in terms of lexical diversity in three different genres. Using both quantitative and qualitative methods, we analyze one statistical and two neural systems, and we compare them to a human reference translation. Our quantitative analyses based on lexical diversity metrics show diverging results; however, translation systems, particularly neural ones, mostly exhibit larger lexical diversity than their human counterparts. Nevertheless, a qualitative method shows that these quantitative results are not always a reliable tool to assess true lexical diversity and that a lot of lexical “creativity”, especially by neural translation systems, is often unreliable, inconsistent, and misguided.

• Jezik: Angleški jezik
• Ključne besede: machine translation, neural translation systems, lexical diversity, type-token ratio, measure of textual lexical diversity
• Vrsta gradiva: Članek v reviji
• Tipologija: 1.01 - Izvirni znanstveni članek
• Organizacija: FF - Filozofska fakulteta
• Status publikacije: Objavljeno
• Različica publikacije: Objavljena publikacija
• Leto izida: 2022
• Št. strani: 14 str.
• Številčenje: Vol. 13, iss. 2, art. 93
• PID: 20.500.12556/RUL-137294
• UDK: 81\'25\'322.4
• ISSN pri članku: 2078-2489
• DOI: 10.3390/info13020093
• COBISS.SI-ID: 100548099
• Datum objave v RUL: 09.06.2022

Gradivo je del revije

• Naslov: Information
• Skrajšan naslov: Information
• Založnik: MDPI
• ISSN: 2078-2489
• COBISS.SI-ID: 18497046

Licence

• Licenca: CC BY 4.0, Creative Commons Priznanje avtorstva 4.0 Mednarodna
• Povezava: http://creativecommons.org/licenses/by/4.0/deed.sl
• Opis: To je standardna licenca Creative Commons, ki daje uporabnikom največ možnosti za nadaljnjo uporabo dela, pri čemer morajo navesti avtorja.
• Začetek licenciranja: 15.02.2022

Sekundarni jezik

• Jezik: Slovenski jezik
• Ključne besede: strojno prevajanje, nevronski prevajalniki, leksikalna diverziteta, razmerje različnic in pojavnic, merjenje besedilne leksikalne diverzitete

Projekti

• Financer: ARRS - Agencija za raziskovalno dejavnost Republike Slovenije
• Številka projekta: P6-0215
• Naslov: Slovenski jezik - bazične, kontrastivne in aplikativne raziskave