Nice, I didn’t realize you were a fan of the heuristics and biases literature. I recommend you read George Loewenstein and Dan Ariely if you haven’t already.
I disagree in the last paragraph. Diversity is essential to predict tastes. Having a “delta function” in the data (ratings -> correlation etc.) means no information, hence no prediction power. A broad distribution of tastes always leads to better results. To predict blockbusters is an easy task, but not very useful. I claim: collaborative filtering works because of divergent tastes. The more divergent the better.
Nice, I didn’t realize you were a fan of the heuristics and biases literature. I recommend you read George Loewenstein and Dan Ariely if you haven’t already.
http://sds.hss.cmu.edu/src/faculty/loewenstein.php
http://web.mit.edu/ariely/www/MIT/papers.shtml
I disagree in the last paragraph. Diversity is essential to predict tastes. Having a “delta function” in the data (ratings -> correlation etc.) means no information, hence no prediction power. A broad distribution of tastes always leads to better results. To predict blockbusters is an easy task, but not very useful. I claim: collaborative filtering works because of divergent tastes. The more divergent the better.