Information Density and Quality Estimation Features as Translationese Indicators for Human Translation Classification

Rubino, Raphaël; Lapshinova-Koltunski, Ekaterina; Genabith, Josef van

doi:10.18653/v1/n16-1110

Cited by 22 publications

(28 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Their main aim is to show that the usage of translation corpora in machine translation should be treated with caution, as human translations do not necessarily correspond to the quality standards that non-translated texts have. Rubino et al (2016) use features derived from machine translation quality estimation to clas-sify translations and non-translations motivating their work by the fact that automatic distinction between originals and machine translations was shown to correlate with the quality of the machine translated texts (Aharoni et al, 2014). However, their data does not contain human quality evaluation.…”

Section: Translation Features and Quality Estimationmentioning

confidence: 99%

Translationese Features as Indicators of Quality in English-Russian Human Translation

Kunilovskaya¹,

Lapshinova-Koltunski²

2019

Proceedings of the Second Workshop Human-Informed Translation and Interpreting Technology Associated With RANLP 2019

Self Cite

View full text Add to dashboard Cite

We use a range of morpho-syntactic features inspired by research in register studies (e.g. Biber, 1995;Neumann, 2013) and translation studies (e.g. Ilisei et al., 2010;Zanettin, 2013;Kunilovskaya and Kutuzov, 2018) to reveal the association between translationese and human translation quality. Translationese is understood as any statistical deviations of translations from non-translations (Baker, 1993) and is assumed to affect the fluency of translations, rendering them foreign-sounding and clumsy of wording and structure. This connection is often posited or implied in the studies of translationese or translational varieties (De Sutter et al., 2017), but is rarely directly tested. Our 45 features include frequencies of selected morphological forms and categories, some types of syntactic structures and relations, as well as several overall text measures extracted from Universal Dependencies annotation. The research corpora include English-to-Russian professional and student translations of informational or argumentative newspaper texts and a comparable corpus of non-translated Russian.Our results indicate lack of direct association between translationese and quality in our data: while our features distinguish translations and non-translations with the near perfect accuracy, the performance of the same algorithm on the quality classes barely exceeds the chance level.

show abstract

Section: Translation Features and Quality Estimationmentioning

confidence: 99%

Translationese Features as Indicators of Quality in English-Russian Human Translation

Kunilovskaya¹,

Lapshinova-Koltunski²

2019

Proceedings of the Second Workshop Human-Informed Translation and Interpreting Technology Associated With RANLP 2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…Translationese seems to affect the semantic as well as the structural level of text, but much of its effects can be seen in syntax and grammar (Santos, 1995;Puurtinen, 2003). An interesting aspect of translationese is that, while it is somewhat difficult to detect for the human eye (Tirkkonen-Condit, 2002), it can be machine learned with high accuracy (Baroni and Bernardini, 2006;Rubino et al, 2016). Many ways to automatically detect translationese have been devised, both with respect to textual translations and simultaneous interpreting (Baroni and Bernardini, 2006;Ilisei et al, 2010;Popescu, 2011).…”

Section: Related Workmentioning

confidence: 99%

Proceedings of the 17th International Conference on Spoken Language Translation

Federico¹,

Waibel²,

Knight³

et al. 2020

View full text Add to dashboard Cite

The conference chairs and organizers would like to express their gratitude to everyone who contributed and supported IWSLT. Our IWSLT-20 program exceeds all our expectations in quality and breath, particularly when considering the challenges during a pandemic under lock-downs and health and travel restrictions. We thank the challenge track chairs, organizers, and participants, the program chairs and committee members, as well as all the authors that went the extra mile to submit system and research papers to IWSLT, and make this year's conference our most vibrant than ever. We also wish to express our sincere gratitude to ACL for hosting our conference and for arranging the logistics and infrastructure that allow us to hold IWSLT 2020 as a virtual online conference.

show abstract

“…Such corpora are also helpful as a source of data for teaching material design. There are but few attempts to approach learner translator texts as a manifestation of some variety of translational language or third code (Frawley, 1984;Rubino et al, 2016). If we can pinpoint quantitative tendencies in learner translator's output that run counter to professional translations and non-translations in the target language and understand why they occur, we can target them in the curriculum and raise learner's awareness of real-life issues with target text quality.…”

Section: Related Work: Translationese and Learner Translator Corporamentioning

confidence: 99%

“…Granger and Rayson (1998) suggested that "one way of characterizing a language variety is by drawing up a word category" (p. 121). Rayson et al (2008) developed an implementation of a corpus comparison technique known as keyword analysis, where keywords were understood as words statistically characteristic of one corpus as compared to the other. Explaining the method, the authors stated that it "can be used to discover key words in the corpora, which differentiate one corpus from another; for example, to determine significant patterns of over-or under-use" (p.2).…”

Section: Frequency Distribution Of Word Classesmentioning

confidence: 99%

Learner vs. professional translations into Russian: Lexical profiles

2018

View full text Add to dashboard Cite

One of the relatively recent trends in learner corpora research is building and exploiting learner translator corpora. Within corpus-based translation studies (CTS) translations are approached as a special variety of the target language. They are usually represented by texts produced by professional translators and are studied as manifestations of the current translational norm. Learner translations can be seen as a more specific variant of the said variety, which is likely to deviate from the accepted translational norm. As of now, typical linguistic features of learner translations as opposed to professional ones are only tentatively described. We hypothesize that these texts should demonstrate heavier translationese features due to the lack of professional translational skills, comparatively poor source language processing competence and target language production skills. The aim of this research is to compare learner and professional Russian translations of English mass-media texts with the reference Russian corpus of non-translations to reveal lexical differences between the three. We found that learner translations consistently showed more distance from non-translations than their professional counterparts, while both learner and professional translations undoubtedly had discursive features which made them linguistically different from naturally occurring language. These findings might help define (non)professionalism in translation and shed light on correlation between the linguistic features of a given text and translation quality, as well as contribute to pedagogical approaches to translator education.

show abstract

Information Density and Quality Estimation Features as Translationese Indicators for Human Translation Classification

Cited by 22 publications

References 28 publications

Translationese Features as Indicators of Quality in English-Russian Human Translation

Translationese Features as Indicators of Quality in English-Russian Human Translation

Proceedings of the 17th International Conference on Spoken Language Translation

Learner vs. professional translations into Russian: Lexical profiles

Contact Info

Product

Resources

About