On Fri, Jul 15, 2022 at 07:10:55PM +0000, Andrew M.A. Cater wrote: > Debconf is on in Kosovo right now. If I had to work out Albanian > gender mappings from names, I'd have no clue.
I decided to take a random name: the President of Kosovo. In Wikipedia I see it is Vjosa Osmani, a name completely unfamiliar to me. >>> print(d.get_gender(u"Vjosa")) female This time the guess happened to be correct. > > Then S. Indian - Malayalam character sets?? and names from a number of > Indian languages then Israel and Hebrew/Arabic > Taiwan had Chinese character sets and names I tried various Hebrew names: names in Hebrew letters are always unknown. My name (a rare one) is unknown even in Latin letters. However some common Hebrew names in Latin letters are detected correctly. I have not tried to do any proper sampling or experiment. Just a random data point. -- mail / xmpp / matrix: tzaf...@cohens.org.il