Netflix $1mm Prize For Improving Their Recommendation Engine Still Up For Grabs
Crowd Sourcing their recommendation engine / algorithm (which frankly needs a lot more then 10% improvement) is a great idea, and well worth the $1mm USD Netflix is offering to the team or group that pulls it off. I'm shocked they are struggling so much with this as I've never been that impressed w/ the accuracy or results of their system; maybe i underestimated the complexity of the problem - or maybe they should reconsider their approach and go the route of last.fm i.e. tagging or something like Pandora and the music genome project... certainly features like tags would only enhance their system and engage their audience more, and like last.fm, i might like something you do that no algorithm would otherwise find.
Last October, Netflix, the online movie rental service, announced that it would award $1 million to the first person or team who can devise a system that is 10 percent more accurate than the company’s current system for recommending movies that customers would like.
About 18,000 teams from more than 150 countries — using ideas from machine learning, neural networks, collaborative filtering and data mining — have submitted more than 12,000 sets of guesses. And the improvement level to Netflix’s rating system is now at 7.42 percent.
The competition is “three-quarters of the way there in three-quarters of a year,” said Reed Hastings, the chairman and chief executive of Netflix, based in Los Gatos, Calif.
I wonder though, who owns the rights to the work? Anyone know?