Newswise — Recommender systems help consumers by selecting products they will probably like and might buy and help retailers boost sales. They are a huge and growing business. Two of the pioneers of recommender systems deconstruct the current recommenders, help you understand what Amazon, Netflix, and other popular recommenders are doing behind the scenes, and describe new algorithms coming into use that better understand user preferences. The article covers collaborative recommenders, personalized and nonpersonalized recommenders, and dimensionality reduction algorithms. It explains how recommenders are tested and evaluated, and tackles the tricky issue of privacy.