Even though it’s a contest that most readers likely don’t realize is underway, a team may have won the $1 million Netflix Prize aimed at improving the company’s movie recommendation algorithm.
Five teams on the leaderboard saw at least 9.50 percent improvement, though just one team, BellKor’s Pragmatic Chaos, was able to reach the necessary 10 percent improvement level. To clarify, BellKor’s Pragmatic Chaos is a group of four teams — comprised of computer engineers and statisticians — working from four countries who claimed the prize.
All other contestants have 30 days to try and top the group’s 10.05 percent rating, or they’ll get the $1 million prize.
The contest started in 2006 and features entries from more than 50,000 individuals and teams of contestants. Netflix saw Team Pragmatic Theory and Team Bellkor in Chaos leading the contest, and allowed both of them to join forces — allowing the two groups to combine forces and push over the magic 10 percent threshold.
Netflix uses its Cinematch software to help recommend movies for subscribers, depending on which movies they rent, watch, and rate. Although the algorithm usually recommends similar movies, there are times when it will recommend a movie or TV series that subscribers have absolutely zero interest in.
I applaud Netflix for opening up such a contest, as the Netflix subscribers ultimately win when we are able to receive better movie recommendations.