Very cool, but I'd be interested to read the actual paper. The way their methodology was described, it seems like it'd be vulnerable to overfitting. http://en

Very cool, but I'd be interested to read the actual paper. The way their methodology was described, it seems like it'd be vulnerable to overfitting. http://en.wikipedia.org/wiki/Overfitting http://en.wikipedia.org/wiki/Overfitting