Deep Poisson Factorization Machines: factor analysis for mapping behaviors in journalist ecosystem
This work provides a method for analyzing journalist behaviors at an individual level, which could help untangle complex interactions in newsrooms, though it appears incremental with extensions to existing factorization techniques.
The authors tackled the problem of understanding journalist behaviors in online ecosystems by proposing Poisson Factorization Machines (PFM), a Bayesian model that extends Poisson Matrix Factorization to incorporate temporal interactions and label information, with results including efficient inference procedures and a deep architecture.
Newsroom in online ecosystem is difficult to untangle. With prevalence of social media, interactions between journalists and individuals become visible, but lack of understanding to inner processing of information feedback loop in public sphere leave most journalists baffled. Can we provide an organized view to characterize journalist behaviors on individual level to know better of the ecosystem? To this end, I propose Poisson Factorization Machine (PFM), a Bayesian analogue to matrix factorization that assumes Poisson distribution for generative process. The model generalizes recent studies on Poisson Matrix Factorization to account temporal interaction which involves tensor-like structure, and label information. Two inference procedures are designed, one based on batch variational EM and another stochastic variational inference scheme that efficiently scales with data size. An important novelty in this note is that I show how to stack layers of PFM to introduce a deep architecture. This work discusses some potential results applying the model and explains how such latent factors may be useful for analyzing latent behaviors for data exploration.