Large-scale, Language-agnostic Discourse Classification of Tweets During COVID-19

TL;DR

This paper introduces a language-agnostic method for classifying Twitter discourse during COVID-19, enabling large-scale, efficient analysis of public attention across multiple languages.

Contribution

It presents a novel approach using language-agnostic tweet representations for scalable discourse classification during crises.

Findings

Feasible large-scale surveillance of COVID-19 discourse

Effective lightweight classifiers with out-of-the-box representations

Analyzed over 26 million tweets during the pandemic

Abstract

Quantifying the characteristics of public attention is an essential prerequisite for appropriate crisis management during severe events such as pandemics. For this purpose, we propose language-agnostic tweet representations to perform large-scale Twitter discourse classification with machine learning. Our analysis on more than 26 million COVID-19 tweets shows that large-scale surveillance of public discourse is feasible with computationally lightweight classifiers by out-of-the-box utilization of these representations.