On Tue, Nov 27, 2018 at 6:54 AM Rebecca Petras
<[hidden email]> wrote:
> Over the past five years, we have worked with various Wikimedia communities focused on simplified health articles being translated into non-English Wikipedia sites. The biggest area where we have focused is east Africa, especially for Swahili. We would like to expand this work, as it is our mission to increase the ability to get critical information openly in marginalized languages. However, we really feel we need to ask two main questions:
> When translating medical articles from English to Swahili, is the translation of higher quality if it is simplified first?
Can you expand what you mean? I am not sure if I understand your question.
> Can a health-specific machine translation engine, used by community who are translating articles, increase the ability to translate medical articles while maintaining quality?
The quality of many of the standard machine translation engines will
rely on the source and destination languages you choose. While some
language pairs give high quality results, some may not. Which machine
translation tool do you have in mind? And for what language pair?
Also, do you use Content Translation?
> I am would like to know your thoughts on whether a research project addressing one or both of those questions would be acceptable – they are key questions for our community.
Any research project that can address the needs of your community can
be great. :) If you expand some more what you want to use the results
of this line of research for, we can try to help you with figuring out
whether the answer to some of your questions is already in the
literature, and through that save some of your time. :)
> Thanks very much,
> Rebecca Petras
> Deputy Director
> Translators without Borders