Looking at Recommender Systems

IEEE deconstructs and compares recommender systems.   " ... By now, online shoppers are accustomed to getting personalized suggestions from the sites they visit. Amazon compiles lists of books we might like. Netflix suggests videos to watch. And Pandora builds personalized music streams by predicting what we’ll want to listen to. Recommenders, which started as relatively crude and often inaccurate predictors of behavior, are now extremely sophisticated systems that often seem to know your tastes better than you do. But how do recommenders work, exactly? Two researchers who’ve been building and analyzing these systems for nearly as long as they’ve been in existence explain how companies size you up and try to match your tastes to those of others. .... "