Dynamic Poisson Factorization

Models for recommender systems use latent factors to explain the preferences and behaviors of users with respect to a set of items (e.g., movies, books, academic papers). Typically, the latent factors are assumed to be static and, given these factors, the observed pref-erences and behaviors of users are assumed to be generated without order. These assumptions limit the explorative and predictive capa-bilities of such models, since users ’ interests and item popularity may evolve over time. To address this, we propose dPF, a dynamic matrix factorization model based on the recent Poisson factoriza-tion model for recommendations. dPF models the time evolving latent factors with a Kalman filter and the actions with Poisson dis-tributions. We derive a scalable variational inference algorithm to infer the latent factors. Finally, we demonstrate dPF on 10 years of user click data from arXiv.org, one of the largest repository of scientific papers and a formidable source of information about the behavior of scientists. Empirically we show performance improve-ment over both static and, more recently proposed, dynamic rec-ommendation models. We also provide a thorough exploration of the inferred posteriors over the latent variables. 1