Netflix Prize probably won

[ultrafilter]

About three years ago, Netflix posted a contest: improve on their ratings prediction algorithm by at least 10%, and win a million dollars. Last Friday, a team of researchers from around the world announced that they’ve met the threshold. If nobody posts a better algorithm by July 26th, the contest will be closed.

Those of you who are interested in machine learning and data mining will find this interesting in its own right. Those who aren’t, you’ll get better recommendations from Netflix in the near future.

The NY Times had this to say:

After nearly three years and entries from more than 50,000 contestants, a multinational team says that it has met the requirements to win the million-dollar Netflix Prize: It developed powerful algorithms that improve the movie recommendations made by Netflix’s existing software by more than 10 percent.

The online movie rental service uses its Cinematch software to analyze each customer’s film-viewing habits and recommends other movies that customer might enjoy. Because accurate recommendations increase Netflix’s appeal to its customers, the movie rental company started a contest in October 2006, offering $1 million to the first contestant that could improve the predictions by at least 10 percent.

…

On Friday, a coalition of four teams calling itself BellKor’s Pragmatic Chaos — made up of statisticians, machine learning experts and computer engineers from America, Austria, Canada and Israel — declared that it had produced a program that improves the accuracy of the predictions by 10.05 percent.

Under the rules of the contest, Netflix said that other contestants now have 30 days to try to do even better. If they cannot, BellKor’s Pragmatic Chaos will collect the $1 million.

[otorophile]

Could the methods used be readily applied to other areas or situations?

[ultrafilter]

Almost certainly, although we can’t say too much more until we know what they are.