Reproduce “SEISMIC: A Self-Exciting Point Process Model for Predicting Tweet Popularity” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Qingyuan Zhao, Murat A. Erdogdu, Hera Y. He, Anand Rajaraman, Jure Leskovec
SEISMIC predicts a post’s eventual reshare count by modeling time-varying resharing likelihood, follower exposure from earlier resharers, and reaction delays. It estimates cascade growth with a branching-process model and training-calibrated corrections. On tweets with at least 50 retweets, its median absolute percentage error after one hour is 15%, using retweets within 14 days as the final-count reference; it cannot reliably predict supercritical cascades.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.