Could Corpus Linguistics help users to make better translations?
I forget what prompted me to ask this question but it probably has to do with translating "fixed expressions". Currently, potentially hundreds if not more beginner users are making similar mistakes translating and it is only by more knowledgeable users to correct these common errors. Could the use of Corpus linguistics help users to create more accurate translations by looking similar uses of phrases in different contexts in the foreign language? Or because there is a new growing body of translated documents on this site, could Duolingo could create its own bilingual/multilingual corpora?
For those unfamiliar with a Corpora system, here is an example: http://corpus.byu.edu/