KDnuggets : News : 2006 : n19 : item1 NEXT >

Features

From: Jim Bennett
Date: Oct 2, 2006
Subject: Netflix $1 million prize for improving accuracy of movie recommendations

Netflix is pleased to announce the Netflix Prize, an award of $1 million to the first person or team who can achieve certain accuracy goals when recommending movies based on personal preferences. The company also made 100 million anonymous movie ratings available to contestants. Complete details for registering and competing for the Netflix Prize are available at

www.netflixprize.com.

We especially invite members of the data mining, statistics and machine learning communities to participate.

[Editor Note: On Oct 9, one week after the announcement, there were already 9940 contestants on 8152 teams from 99 different countries. There were 82 valid submissions.

Two teams have already beaten Netflix own performance, and one qualified for 50,000 Progress prize

see www.netflixprize.com/leaderboard

[Note: here is a news story on the prize - Editor]

Oct 2, 2006. LOS ANGELES (Reuters) - Online DVD rental company Netflix Inc (Nasdaq:NFLX - news) 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. ADVERTISEMENT

Netflix's Web-based recommendation system "learns" what kinds of films subscribers like by asking them to rate the films they watch. The system then recommends lists of similar titles, unique to each user.

Netflix Chairman and Chief Executive Reed Hastings, who made his fortune designing software, predicted that Web-based recommendation systems would play "an increasingly significant commercial role in the future."

The winning software designer must improve the accuracy of Netflix's current rating system by 10 percent. The system is composed of more than 1 billion ratings, in which subscribers use one to five stars to describe how much they liked a film.

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.

"If we can guess the really great ones for you and the ones you really abhor ... we can really choose the four- and five-star movies that people want," Bennett 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.

Read more.


KDnuggets : News : 2006 : n19 : item1 NEXT >

Copyright © 2006 KDnuggets.   Subscribe to KDnuggets News!