The NYT built an interactive dialect test - it got me spot-on: strong bias to upstate New York. http://nyti.ms/1cmDo5a
-
-
@avibryant@mat_kelcey@ryanoneill With 25 questions, most of which aren't binary, we seem to be using way more bits than necessary. -
@johnmyleswhite@mat_kelcey@ryanoneill and is it always the same 25? You want a tree of questions with minimal average depth. -
@avibryant@mat_kelcey@ryanoneill I believe it's always the same 25. Certainly@jacqui and I seemed to have answered same questions. -
@johnmyleswhite@avibryant@mat_kelcey@ryanoneill@jacqui You could calculate the conditional entropy to display optimal sequences. -
@johnmyleswhite@avibryant@mat_kelcey@ryanoneill@jacqui I did something almost exactly like that for an online troubleshooter in 2000
End of conversation
New conversation -
-
-
@avibryant@mat_kelcey@ryanoneill I think the next@kaggle contest should be the construction of a minimal set of predictors for this.Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.