Recent posts
Netflix Prize
The Netflix Prize was the most reputed open competition for the best collaborative filtering algorithms that could predict user ratings for films based on previous ratings. Gravity's team consistently performed among the top 10, and finished (tied) with the highest performance in the competition. The competition was announced in late 2006, after the significant benefits of quality and personalized recommendations became apparent. For Netflix, just a 10% increase in recommendation quality meant an approximately USD 80-90 million additional revenue due to increased customer life-time value and subscription upsellings.The prize would only be awarded to a team that achieved a 10%+ improvement over Netflix's own Cinematch algorithm.
The competition showed that even a small change in prediction accuracy can greatly influence the quality of recommendations and the related business benefits. The competition was announced because algorithms used at that time were not sophisticated enough to yield significant benefits to businesses, and as a result, a new generation of recommendation algorithms emerged.
Gravity & The Ensemble Team were able to outperform the Netflix recommendation system by 10.1% and were consistently in the Top 10 throughout the competition, which hosted over 40,000 participants.
For More Details of Gravity's team participation at the Netflix Prize Competition.