An Interpretable GeoAI Framework for Longitudinal Spatiotemporal Analysis of Digital Discourse — Replication Files

This repository contains anonymized replication materials for the study “An Interpretable GeoAI Framework for Longitudinal Spatiotemporal Analysis of Digital Discourse.” It includes fully documented code, data dictionaries, and aggregated model outputs required to reproduce all analyses reported in the manuscript. Raw geotagged social media data are not redistributed due to privacy and data-use restrictions; access to the original dataset is available via the Harvard Center for Geographic Analysis (CGA) GeoTweet Archive v2.0, subject to institutional approval.