RusLTC

Supported by the Russian Foundation for Basic Research (RFBR/RFFI), grant No. 17-06-00107 (2017–2019).

Linguistic properties of translations: descriptive norms and translation quality estimation (based on English-to-Russian mass-media texts)

The project set out to identify translationese indicators for English-to-Russian translation: human-interpretable linguistic features that distinguish translations from comparable texts originally written in the target language. We built a system of corpus resources for studying translation quality and translation varieties, with the publicly available Russian Learner Translator Corpus (RusLTC) at its core, and developed a method for selecting functionally comparable texts from heterogeneous collections. A set of 45 morphosyntactic and lexical features captures translationese reliably (92% classification accuracy for Russian and 83% for German) and shows that student translations are more translationese than professional ones. However, these features correlate only weakly with human judgements of translation quality.

Read the full popular-scientific description

RusLTC homepage

Interview (in Russian)

Алетейя | Мария Куниловская «Лингвистические свойства переводов», Tyumen State University

Some publications on the project

• Kunilovskaya, M., & Kutuzov, A. (2018). Universal Dependencies-based syntactic features in detecting human translation varieties. In Proceedings of the 16th International Workshop on Treebanks and Linguistic Theories (TLT16), 27–36. ACL.
• Kunilovskaya, M., Morgoun, N., & Pariy, A. (2018). Learner vs. professional translations into Russian: Lexical profiles. Translation & Interpreting, 10(1), 33–52. DOI: 10.12807/ti.110201.2018.a03
• Kunilovskaya, M. (2017). Frequency distribution and usage of discourse markers in English to Russian translation: learner vs. professional translational behavior. Corpus Linguistics 2017, University of Birmingham.
• Kunilovskaya, M., & Kutuzov, A. (2017). Testing target text fluency: A machine learning approach to detecting syntactic translationese in English-Russian translation. In K. Menzel, E. Lapshinova-Koltunski, & K. Kunz (Eds.), New perspectives on cohesion and coherence: Implications for translation, 75–104. Language Science Press. DOI: 10.5281/zenodo.814466
• Kunilovskaya, M. (2017). Linguistic tendencies in English to Russian translation: the case of connectives. In Computational Linguistics and Intellectual Technologies: Proceedings of the International Conference “Dialogue 2017”, vol. 2, 221–233.
• Kunilovskaya, M. A., Ilyushchenya, T. A., & Kovyazina, M. A. (2017). Оценка качества учебного перевода на основе разметки по ошибкам: корпусный подход [Translation quality assessment based on error annotation: a corpus approach]. Вестник Тюменского государственного университета. Гуманитарные исследования. Humanitates, 3(3), 94–112. DOI: 10.21684/2411-197X-2017-3-3-94-112