The Netflix Prize Netflix Netflix is a subscriptionbased
The Netflix Prize
Netflix �Netflix is a subscription-based movie and television show rental service that offers media to subscribers: �Physically by mail �Over the internet �Has a catalog of over 100, 000 movies and television shows �Subscriber base of over 10 million
Recommendations �Netflix offers users the ability to rate movies and television shows that they have seen �Depending on those ratings, Netflix provides recommendations of movies and television shows that the subscriber would like to watch �These recommendations are based on an algorithm called Cinematch
Cinematch �Uses straightforward statistical linear models with a lot of data conditioning �This means that the more a subscriber rates, the more accurate the recommendations will become
Netflix Prize �Competition for the best collaborative filtering algorithm to predict user ratings for movies and television shows, based on previous ratings �Offered a $1 million prize to the team who could improve Cinematch’s accuracy by 10% �Awarded a $50, 000 progress prize for the team who makes the most progress for each year before the 10% mark was reached �The contest started on October 2, 2006 and would run until at least October 2, 2011, depending on when a winner was chosen
Metrics �The accuracy of the algorithms was measured by using root mean square error, or RMSE �Chosen because it is a well-known, single value that can account for and amplify the contributions of errors such as false positives and false negatives
Metrics �Cinematch scored 0. 9525 on the test subset �The winning team needed to score at least 10% lower, with an RMSE of 0. 8563
Results �The contest ended on June 26, 2009 �The threshold was broken by the teams “Bell. Kor's Pragmatic Chaos” and “The Ensemble”, both achieving a 10. 06% improvement over Cinematch, with an RMSE of 0. 8567 �“Bell. Kor's Pragmatic Chaos” won the prize due to the team submitting their results 20 minutes before “The Ensemble”
Netflix Prize Sequel �Due to the success of their contest, Netflix announced another contest to further improve their recommender system �Unfortunately, it was discovered that the anonymized customer data that they provided to the contestants could actually be used to identify individual customers �This, combined with a resulting investigation by the FTC and a lawsuit, led Netflix to cancel their sequel
Sources �http: //blog. netflix. com/2010/03/this-is-neil-hunt-chief-productofficer. html �http: //www. netflixprize. com �http: //www. nytimes. com/2010/03/13/technology/13 netflix. html? _r=1
- Slides: 10