Read the release here.

Original Article: Netflix quietly announced a new contest recently. The announcement was found in the forum at the NetflixPrize site, which hosted the first contest.

If you are unfamiliar with the Netflix Prize contest, it has been an ongoing initiative over the last several years in which Netflix promised to reward the person who improved the service’s recommendation algorithm by 10%, with a million dollars.

That contest came to a close last month. The winners will be announced in September. Now comes part 2. Netflix Chief Product Officer Neil Hunt writes:

The advances spurred by the Netflix Prize have so impressed us that we’re planning Netflix Prize 2, a new big money contest with some new twists.

While the first contest has been remarkable, we think Netflix Prize 2 will be more challenging, more fun, and even more useful to the field.

Hunt promises more details when the winners of the first prize are announced. Judging by responses in the forum, Netflix will have no problem drawing plenty of enthusiasm for the next contest. This should ultimately make Netflix a better service for its users. The goals are, after all, aimed at improving how it works.

"The contest made available to contestants 100 million anonymous movie ratings ranging from one to five stars, the largest such data set ever released," engineer Jon Sanders explains on the company blog. "All personal information identifying individual Netflix members was removed from the prize data, which contained only movie titles, star ratings and dates but no text reviews."

Netflix has not confirmed the winner of the prize yet. They are waiting to validate the submissions. There were about 44,000 entries from 5,169 teams in 186 countries. Not a bad competition.

"The winning team will take home $1 million and Netflix members will benefit as we will incorporate the winner’s technology in our recommendations system to continually improve the movie recommendations we make for you," says Sanders.

Last week Netflix announced its financial results for the second quarter. The company reported a 22% increase in profit thanks to subscriber growth. Earlier this year, the company reached a couple of milestones – surpassing 10 million subscribers and shipping its second billionth movie (which happened much faster than the first billionth).

The physics arXiv blog noted Netflix claimed to have removed personal details from the dataset before making it available. However, Arvind Narayanan and Vitaly Shmatikov at the the University of Texas at Austin figured out how to de-anonymize that data.

The research paper on how they did it demonstrated the inherent risk in publishing such micro-data, or information about specific individuals.

“Using the Internet Movie Database as the source of background knowledge, we successfully identified the Netflix records of known users, uncovering their apparent political preferences and other potentially sensitive information,” the researchers said in the paper’s abstract.

Through their algorithmic work, the researchers could tie information in the Netflix dataset with recommendations made on the Internet Movie Database website:

We expect that for Netflix subscribers who use IMDb, there is a strong correlation between their private Netflix ratings and their public IMDb ratings. Note that our attack does not require that all movies rated by the subscriber in the Netflix system be also rated in IMDb, or vice versa. In many cases, even a handful of movies that are rated by the subscriber in both services would be sufficient to identify his or her record in the Netflix Prize dataset…

Briefly, people who rated movies publicly around the same time they rated those movies privately gave the researchers enough data to figure out details about one person.

“A natural question to ask is why would someone who rates movies on IMDb – often under his or her real name – care about privacy of his movie ratings?” the researchers asked.

“Consider the information that we have been able to deduce by locating one of these users

The best of these efforts, as they stand now, will only qualify for the $50,000 “Progress Prize.” The $1 million is reserved for the first team to achieve a 10% improvement, and the Leaderboard currently tops out with The Thought Gang’s 1.06% tweak.

The contest appears to be drawing quite a crowd, though. “There currently 10208 contestants on 8381 teams from 99 different countries,” according to the Leaderboard. Things seem to be picking up, as well – Netflix has “received 92 valid submissions; 26 in the last 24 hours.”

It seemed like a good time to check in again on the story. Several articles appeared yesterday to announce WXYZ Consulting’s 0.9430% improvement, and although that record has already fallen by the wayside, the pieces still offer insight into how the Netflix Prize is regarded.

Jim Bowery posted on Slashdot, and wrote, “This is pretty impressive given the previously quoted researcher who said: You’re competing with 15 years of really smart people banging away at the problem.’”

Techdirt’s Mike admitted to “coming around.” “A week ago, we weren’t sure if there really were that many people who would be willing to help Netflix improve their recommendation engine for just the chance at winning a million dollars . . . . Apparently, we severely underestimated the competitive nature of Netflix fans.”

Skeptics still exist, though. “Spartacus” responded to Mike’s article, writing, “They . . . are not even a tenth of the way to the grand prize of 1 million dollars. So I don’t think this is that mind blowing yet.” That’s fair. I also like the use of the word “yet.”

Tag:

Add to Del.icio.us | Digg | Yahoo! My Web | Furl

Bookmark WebProNews:

Doug is a staff writer for WebProNews. Visit WebProNews for the latest eBusiness news.