In June, 2009 NI TEKNA Team has submitted a solution for the Netflix Prize competition . The Netflix Prize sought to substantially improve the accuracy of predictions about how much someone is going to enjoy a movie based on their movie preferences.
We did not win the million dollar prize, but however our solution achieves significant performance improvements of the initial recommender system that was used by Netflix before the competition started. Our participation on this competition was a great experience for us, because we have mastered the most common algorithms used for recommender systems (like collaborative filtering, singular value decomposition, etc) and have successfully implemented parallel and distributed processing of a very large database. By implementing parrallel processing in multiple theads, and introducing some other optimizations, we have speeded up the processing of the data that would normally need few years to complete, so it finishes in half a day without making any hardware upgrades.