A Google–Wikipedia–Twitter Model as a Leading Indicator of the Numbers of Coronavirus Deaths

• DOI: http://doi.org/10.1002/isaf.1482
• Date: 01 July 2020
• Author: Veda C. Storey,Daniel E. O'Leary
• Published date: 01 July 2020

RESEARCH ARTICLE

A Google–Wikipedia–Twitter Model as a Leading Indicator of

the Numbers of Coronavirus Deaths

Daniel E. O'Leary

1

|Veda C. Storey

2

1

University of Southern California, Los

Angeles, CA, USA

2

Georgia State University, Atlanta, GA, USA

Correspondence

Daniel E. O'Leary, University of Southern

California, Los Angeles, CA, USA.

Email: oleary@usc.edu.

Summary

Forecasting the number of cases and the number of deaths in a pandemic provides

critical information to governments and health officials, as seen in the management

of the coronavirus outbreak. But things change. Thus, there is a constant search for

real-time and leading indicator variables that can provide insights into disease propa-

gation models. Researchers have found that information about social media and sea-

rch engine use can provide insights into the diffusion of flu and other diseases.

Consistent with this finding, we found that a model with the number of Google

searches, Twitter tweets, and Wikipedia page views provides a leading indicator

model of the number of people in the USA who will become infected and die from

the coronavirus. Although we focus on the current coronavirus pandemic, other

recent viruses have threatened pandemics (e.g. severe acute respiratory syndrome).

Since future and existing diseases are likely to follow a similar search for information,

our insights may prove fruitful in dealing with the coronavirus and other such dis-

eases, particularly in the early phases of the disease.

Subject terms: coronavirus, COVID-19, unintentional crowd, Google searches,

Wikipedia page views, Twitter tweets, models of disease diffusion.

1|INTRODUCTION

The coronavirus disease 2019 (COVID-19) is a newly discovered

disease, causing respiratory illness. The disease was first identified

in December 2019, in Wuhan, China, and has since spread globally,

resulting in a pandemic (World Health Organization, 2020). This

paper provides an analysis of data early in the coronavirus life cycle,

with the data-gathering process ending on 5 April 2020. Although

the pandemic is global, our focus is primarily on the USA and

English language sources of information because, at the time of

writing, the USA had just become the world leader in the number

of active coronavirus cases.

Although researchers have noted the importance of many

aspects of the pandemic (e.g. O'Leary, 2020), forecasting the number

of cases and deaths is, perhaps, one of the most important issues.

Unfortunately, researchers have noted that it has been difficult to

forecast cases of the coronavirus (Cohen, 2020). Further,

Butler (2013) and others have noted the extent to which disease

forecasting models need to change over time. In addition, some

researchers expect more pandemics in the future. As noted by

Haseltine (2020) ‘Once this new epidemic has faded into memory,

will it be five or 10 years until the next one?’As a result, it is impor-

tant to continue to search for, and better understand, variables that

provide a leading indicator of the number of coronavirus cases and

deaths in a range of different settings. As noted by Milinovich, Wil-

liams, Clements, and Hu (2014), we need surveillance systems to

constantly monitor for infectious diseases. This paper does so by

building a model using data from two types of social media

(Wikipedia and Twitter tweets) and Google searches to estimate the

number of coronavirus cases and deaths in the USA early in the life

cycle of the disease. Althouse et al. (2015) referred to these data,

such as Google searches, Twitter posts, Wikipedia access, and other

data streams as ‘novel data streams’(NDSs) and suggested the use of

NDSs in models of public health.

DOI: 10.1002/isaf.1482

Intell Sys Acc Fin Mgmt. 2020;27:151–158.wileyonlinelibrary.com/journal/isaf© 2020 John Wiley & Sons, Ltd.151