David Chudzicki's Blog: speaking of machine learning

Learn From Data

Sunday, December 18, 2011

Umm, not exactly:

And "Albanian" doesn't come first ("Africaans") does, so it's not even that kind of bug.

At December 18, 2011 at 9:42 PM , F said...: Did you try translating it?
At December 18, 2011 at 10:27 PM , David said...: Well, that's embarrassing. (I didn't.)

Your blog is beautiful, by the way.
At December 19, 2011 at 11:35 AM , Jeff Kaufman said...: Maybe it's alphabetically first of the languages that this could be?

Or some of these words (media? frame?) are statistically much more common in albanian than english?
At December 19, 2011 at 3:55 PM , David said...: Presumably it's something like the second one. I would imagine ties go to English for me.
At May 9, 2012 at 3:33 PM , David said...: (Update) Ran into someone who works on Chrome at a contra dance. He says the language recognizer works by the distributions of sequences of three letters (like trigram, but for letters rather than words).