Fast Non-Bayesian Poisson Factorization for Implicit-Feedback Recommendations
From MaRDI portal
Publication:61639
DOI10.48550/ARXIV.1811.01908arXiv1811.01908MaRDI QIDQ61639FDOQ61639
Authors: David Cortes 
Publication date: 5 November 2018
Abstract: This work explores non-negative low-rank matrix factorization based on regularized Poisson models (PF or "Poisson factorization" for short) for recommender systems with implicit-feedback data. The properties of Poisson likelihood allow a shortcut for very fast computations over zero-valued inputs, and oftentimes results in very sparse factors for both users and items. Compared to HPF (a popular Bayesian formulation of the problem with hierarchical priors), the frequentist optimization-based approach presented here tends to produce better top-N recommendations with significantly shorter fitting times, on top of having sparse solutions.
Cited In (1)
This page was built for publication: Fast Non-Bayesian Poisson Factorization for Implicit-Feedback Recommendations
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q61639)
