Abstract— Previous work on recommender systems mainly focus on effitting the ratings provided by users. However, the response patterns, i.e., some items are rated while others not, are generally ignored. We argue that failing to observe such response patterns can lead to biased parameter estimation and sub-optimal model performance. Although several pieces of work have tried to model users’ response patterns, they miss the effiectiveness and in-terpretability of the successful matrix factor-ization collaborative effiltering approaches. To bridge the gap, in this paper, we unify ex-plicit response models and PMF to estab-lish the Response Aware Probabilistic Ma-trix Factorization < Final Year Projects >(RAPMF) framework. Weshow that RAPMF subsumes PMF as a spe-cial case. Empirically we demonstrate themerits of RAPMF from various aspects.