On Tue, Sep 15, 2015 at 11:13 PM, Gerard Meijssen <gerard.meijssen(a)gmail.com
wrote:
One question, when you search for
''Ревест-Сен-Мартен", why did you not
consider every language that uses the Cyrillic script? It is as likely to
find something in Serbian, Macedonian, Belarusian etc ...
Thanks,
GerardM
The rest of the discussion is happening on the phab ticket, but i'll answer
this here. We are using a language detection algorithm that has been
trained against tweets. Tweets are not, on average, as short as the
searches we are detecting the language of but it does an ok job. Trey did a
great job putting together an analysis[1] of this language detection algo.
We will also be using his work there to evaluate other language detection
methods and perhaps change what we are using in the future.
So the short of it is, we chose russian instead of serbian because the
machine learning algorithm said so.
[1]
https://www.mediawiki.org/wiki/User:TJones_(WMF)/Notes/Language_Detection_E…