March 25, 2007
Netflix Users Steadily Competing to Find Better Recommendations for the Rental Service

From time-to-time, I want to initiate the chance to follow up on questions raised in prior posts, particularly looking back at various contests or initiatives that I found to be of interest but have not publicly followed back up on. The first is to return to a major contest launched last year to embrace "the wisdom of the crowd" by Netflix.

Last October, I wrote about Netflix's plan to create the Netflix Prize to award to those who can increase the accuracy of the company's predictions as to what users would like to see.

The company is looking for 10 percent improvement over the accuracy of their own recommendation system.

I wrote, "The amount of intellectual capital that the company may become privy to during this contest demonstrates the power of a collective intelligence, as Henry Jenkins writes about. And, with people saying things like, 'First, I have to generate my test bed and get to work this is so cool. I don't know what it is with me and large, nicely formatted, datasets, but I don't think there's anything that can get me more excited,' they've certainly hit a research nerve with a section of Internet users."

At the time, I questioned, "Could Netflix cause a change in the way companies think about researching complex questions? Or could this be forgotten in a couple of months? We shall see..."

Over at The Netflix Prize site, it looks like the best achievement so far is an improvement of 6.75 percent, submitted today a little after non. This replaced the previous best improvement score of 6.74 percent, which was achieved at 7 a.m. this morning.

In short, of the numbers in the Top 10, all but one happened in the past week, and seven of the 10 happened in the past week. As of right now, user "Gravity" is in the lead, but it looks like there is still a dedicated base of users looking to break the 10 percent improvement mark.

At this point, there are no Grand Prize candidates, since that point has not been reached, but it looks like the community has not yet tired of this contest and that participants continue moving toward an eventual solution to the Netflix challenge.

Netflix admitted that it might take months or years to find an answer to this question but expressed their strong belief in the power of a collective competition to discover a better answer than the company itself could ever come up with, as to how to improve its recommendations. And, if the steady progress of the competition is any indication, it looks like users are well on their way.

For those interested in knowing more about the competition, look here for the FAQ. Looks like this is one initiative that hasn't lost any steam months after I originally wrote about the competition.