Online DVD rental company Netflix Inc on Sunday announced that it would pay $1 million to the first person to develop software to improve the accuracy of Netflix's movie recommendation system by 10 percent.What a cool concept. I like the idea of having a prize rather than paying staff to try and solve this.
The current system comes within one star of accurately predicting a consumer's true feelings about a film, and the company wants to cut that margin by one-quarter of a star, Jim Bennett, vice president of recommendation systems, said.
The prize was modeled on the Longitude Prize, offered by the British government in 1714 to the inventor who could determine a ship's longitude during transoceanic travel.
In the attention economy filtering and recommendation engines are key. We are limited by time, so we can't consume more movies. The only way to improve the service is to get better movies.
One issue I don't know how they can solve is that the mood I am in and the people I watch a movie with affect my enjoyment of it and how I rate it. Without having that information, I don't see how you could always predict how the movie will be rated.
Netflix also needs to change their 1-5 star system to allow people to give .5 star ratings. There are so many times I want to give a movie a 3.5 and have to decide between the 4 and the 3. I bet that would improve the system 10% by itself. Man, I need to go collect my $1 million.
More information at netflixprize.com