Designing weighted and multiplex networks for deep learning user geolocation in Twitter

TL;DR

This paper introduces new deep learning methods using weighted multiplex networks and graph neural networks to improve user geolocation prediction on Twitter, evaluated on US and Latin American datasets.

Contribution

It proposes novel combinations of weighted multigraphs with GNNs and attention mechanisms for geolocation, and provides new datasets for Latin America.

Findings

Enhanced geolocation accuracy over baseline models

Effective integration of user text and network structure

Validation on diverse datasets demonstrates robustness

Abstract

Predicting the geographical location of users of social media like Twitter has found several applications in health surveillance, emergency monitoring, content personalization, and social studies in general. In this work we contribute to the research in this area by designing and evaluating new methods based on the literature of weighted multigraphs combined with state-of-the-art deep learning techniques. The explored methods depart from a similar underlying structure (that of an extended mention and/or follower network) but use different information processing strategies, e.g., information diffusion through transductive and inductive algorithms -- RGCNs and GraphSAGE, respectively -- and node embeddings with Node2vec+. These graphs are then combined with attention mechanisms to incorporate the users' text view into the models. We assess the performance of each of these methods and compare them to baseline models in the publicly available Twitter-US dataset; we also make a new dataset available based on a large Twitter capture in Latin America. Finally, our work discusses the limitations and validity of the comparisons among methods in the context of different label definitions and metrics.