Non-Pharmaceutical Intervention Discovery with Topic Modeling

TL;DR

This paper uses topic modeling on large corpora to automatically identify categories of non-pharmaceutical interventions during COVID-19, reducing the need for manual labeling and aiding pandemic response efforts.

Contribution

It introduces a novel application of topic modeling to discover intervention categories with minimal human effort during a global health crisis.

Findings

Models accurately match human-labeled intervention categories

Reduces manual effort in classifying interventions

Effective on both national and international corpora

Abstract

We consider the task of discovering categories of non-pharmaceutical interventions during the evolving COVID-19 pandemic. We explore topic modeling on two corpora with national and international scope. These models discover existing categories when compared with human intervention labels while reduced human effort needed.