The Netflix Prize was an open computing challenge for the best algorithm to recommend movies to subscribers. The winning team, Bellkor, has improved Netflix’s own cinematch algorithm by 10.05% and will be awarded $1M for their success.

Clearing the hurdle wasn’t easy. Thousands of teams from more than 100 nations tackled the problem for more than three years, sharing their results and algorithms along the way. CEO Reed Hastings said in an interview “You look at the cumulative hours and you’re getting Ph.D.’s for a dollar an hour”. The contest is an example of Prize economics, where competitive incentives are offered as an alternative to in-house research and development. This was a tremendous research bargain for Netflix.

Netflix analyzes customers’ choices and ratings on the movies they have rented and recommends movies in a way that optimizes both the customer’s taste and inventory conditions. Automated recommendation algorithms are seen as a key competitive edge in e-commerce, allowing retailers to guess the types of products and services customers are seeking by looking at their past behavior. For Netflix, a 10% improvement on its algorithms could help move substantially higher numbers of movies and increase customer satisfaction, with a direct boost to profits.

The company’s analytical orientation has already led to a high level of success and growth. Netflix announced it would run a second competition, with shorter time spans of 6 months and 18 months to prize awarding. The new challenge will use demographic data rather than previous ratings as data to base predictions on.

Read More…

Netflix is the world’s largest online movie rental service, with more than 11 million subscribers. It has revolutionized the way people rent movies – by bringing the movies directly to them.