It’s kind of depressing to think that it might work quite well formy project, and beat my snazzy visual search model. But then again, a lot of the statistical tools I’ve come accross are pants. Interesting article though.
ps. For anybody I’ve not told already: my first paper has been accepted. Yay!