# Dynamic Poisson Factorization

arXiv: https://arxiv.org/abs/1509.04640 · HTML: https://mlsanity.com/paper/1509.04640/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 15 Sep 2015.

## Built on by (most cited first, 8 of 8)

- [Towards Psychologically-Grounded Dynamic Preference Models](https://mlsanity.com/paper/2208.01534/) (2022)
- [A Dynamic Meta-Learning Model for Time-Sensitive Cold-Start Recommendations](https://mlsanity.com/paper/2204.00970/) (2022)
- [Modeling Dynamic User Interests: A Neural Matrix Factorization Approach](https://mlsanity.com/paper/2102.06602/) (2021)
- [Evolving Voices Based on Temporal Poisson Factorisation](https://mlsanity.com/paper/2410.18486/) (2024)
- [Universal Multi-Modal Traceformer: Integrating Heterogeneous Context for Process Event Prediction](https://mlsanity.com/paper/2609.24579/) (2026)
- [Tracking Brand-Associated Polarity-Bearing Topics in User Reviews](https://mlsanity.com/paper/2301.07183/) (2023)
- [Scaling up Dynamic Edge Partition Models via Stochastic Gradient MCMC](https://mlsanity.com/paper/2403.00044/) (2024)
- [Structured Neural Marked Point Processes for Interpretable Event Interaction Modeling](https://mlsanity.com/paper/2605.17568/) (2026)

## Abstract (arXiv)

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 preferences and behaviors of users are assumed to be generated without order. These assumptions limit the explorative and predictive capabilities 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 factorization model for recommendations. dPF models the time evolving latent factors with a Kalman filter and the actions with Poisson distributions. 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 improvement over both static and, more recently proposed, dynamic recommendation models. We also provide a thorough exploration of the inferred posteriors over the latent variables.