Active Learning with Bayesian Nonnegative MatrixFactorization for Recommender Systems Contributors: Gönül Aycı Abdullatif Köksal Melih Mutlu Burak Suyunu Prof. Ali Taylan Cemgil Dense Dataset KL Divergence RMSE Random Random Popularity Max Row-Column Max Variance Min Variance Popularity Variance Sparse Dataset KL Divergence RMSE Random Random Popularity Max Row-Column Max Variance Min Variance Popularity Variance