Skip to main content

Bayesian nonparametric poisson factorization for recommendation systems

Author(s): Gopalan, P; Ruiz, FJR; Ranganath, R; Blei, DM

Download
To refer to this page use: http://arks.princeton.edu/ark:/88435/pr1mb7m
Abstract: We develop a Bayesian nonparametric Poisson factorization model for recommendation systems. Poisson factorization implicitly models each user's limited budget of attention (or money) that allows consumption of only a small subset of the available items. In our Bayesian nonparametric variant, the number of latent components is theoretically unbounded and effectively estimated when computing a posterior with observed user behavior data. To approximate the posterior, we develop an efficient variational inference algorithm. It adapts the dimensionality of the latent components to the data, only requires iteration over the user/item pairs that have been rated, and has computational complexity on the same order as for a parametric model with fixed dimensionality. We studied our model and algorithm with large real-world data sets of user-movie preferences. Our model eases the computational burden of searching for the number of latent components and gives better predictive performance than its parametric counterpart.
Publication Date: 1-Jan-2014
Citation: Gopalan, P, Ruiz, FJR, Ranganath, R, Blei, DM. (2014). Bayesian nonparametric poisson factorization for recommendation systems. Journal of Machine Learning Research, 33 (275 - 283)
ISSN: 1532-4435
EISSN: 1533-7928
Pages: 275 - 283
Type of Material: Conference Article
Journal/Proceeding Title: Journal of Machine Learning Research
Version: Final published version. This is an open access article.



Items in OAR@Princeton are protected by copyright, with all rights reserved, unless otherwise indicated.