You are viewing an old version of this page. View the current version.
Compare with Current
View Page History
« Previous
Version 6
Next »
Machine translation quality and post-editing
- Cochrane Russia conducted a study to compare the quality of DeepL, Google Translate and Microsoft Translator machine translation. (The study has not been published yet.)
- Cochrane Germany conducted a study to compare the quality of DeepL machine translation against SYSTRAN machine translation.
- As part of the Cochrane France collaboration with a Université Paris Diderot translation course, Hanna Martikainen published a study on post-editing of neural machine translation in medical texts, and how it impacts on lexico-grammatical patterns in the translations.
- Cochrane Brazil developed training videos for volunteer translators based on the most common mistakes translators were making, and evaluated whether the training videos improved the quality of the volunteers’ translations. The videos are available on YouTube. (The study has not been published yet.)
- As part of the EU-funded INTERACT project, Cochrane members contributed as volunteer writers and readability testers of Cochrane PLS to a PhD on simplifying, reading, and machine translating health content: http://doras.dcu.ie/23124/.
- Also, as part of INTERACT, a case study of the management and training of volunteer translators was conducted at Cochrane Germany.
- Cochrane Croatia conducted a study on the motivation of their volunteer translators, and on whether reminders increase the volunteers’ engagement. (The second study has not been published yet.)
- Results from the INTERACT and Cochrane Croatia volunteer studies were presented in a webinar.
- Cochrane Malaysia started a collaboration with the Universiti Sains Malaysia’s translation department to improve the quality and readability of their translations and to develop a quality framework for Malay.
