Core Data Science


Popularity Prediction for Social Media over Arbitrary Time Horizons

We propose a feature-based approach based on a self-excited Hawkes point process model, which involves prediction of the content’s popularity at one or more reference horizons in tandem with a point predictor of an effective growth parameter that reflects the timescale of popularity growth.


Crowdsourcing with Contextual Uncertainty

We present Theodon, a hierarchical nonparametric Bayesian model, developed and deployed at Meta, that captures both the prevalence of label categories and the accuracy of...


Post approvals in online communities

Moderation in online communities increases the quality of contributions and decreases antisocial behavior, and the tools that online platforms provide are one way that...