# Coronavirus research topics, tracking twenty years of research

## TL;DR

This paper presents a dataset of over 800,000 coronavirus research articles from 2002 to 2024, organized into thematic clusters to help understand research trends and innovations.

## Contribution

The novel contribution is a systematically curated and expert-reviewed dataset of coronavirus research, organized thematically for accessibility and reuse.

## Key findings

- A dataset of over 800,000 coronavirus-related articles was created using natural language processing.
- Research trends were organized into thematic clusters like vaccine development and public health strategies.
- The dataset was reviewed and revised with input from health experts to ensure accuracy and relevance.

## Abstract

Research publications aimed at understanding the various aspects of Coronaviruses, particularly COVID-19, have significantly shaped our knowledge base. While the urgency to monitor COVID-19 in real-time has decreased, the continual influx of new research of monthly articles underscores the importance of systematic review and analysis to deepen our understanding of the pandemic’s broad impact. To explore research trends and innovations in this space, we developed a pipeline using natural language processing techniques. This pipeline systematically catalogues and synthesises the vast array of research articles, leading to the creation of a dataset with more than eight hundred thousand articles from July 2002 to May 2024. This paper describes the content of this dataset and provides the necessary information to make this dataset accessible and reusable for future research. Our approach aggregates and organises global research related to Coronaviruses into thematic clusters such as vaccine development, public health strategies, infection mechanisms, mental health issues, and economic consequences. Also, we have leveraged the contribution of health experts to review and revise the dataset.