An early warning system for future pandemics? | UCI News

Can Twitter be used to spot pandemics before they break out?

To find out, UCI and UCLA researchers are sifting through millions of tweets (and other data) from the months leading up to the major COVID-19 outbreak, looking for anomalies and patterns that would have given early warning of the virus.

“It’s a bit like looking for a needle in a haystack,” admits Andrew Noymer, Associate Professor of Public Health and Disease Prevention at the UCI. “But the stakes are high, so it’s worth trying different approaches.”

The National Science Foundation agreed, awarding the 10-strong UCI-UCLA team nearly $1 million under its new Predictive Intelligence for Pandemic Prevention grant program, which will fund “high-risk, high-reward” research “aimed at reducing the impact.” predict, track and mitigate future pandemics.”

Chen Li, a computer science professor leading the effort at UCI, likens the project to “weather forecasting, where advances in big data technologies and information analysis have led to better predictions that are further away.” A pandemic early detection system could enable “quicker responses from healthcare, medicine and government,” he says.

Li, Noymer and principal investigator Wei Wang, a UCLA professor of computer science and computational medicine, developed the grant proposal last year. Noting that infectious diseases are “sociobiological phenomena, leaving both social and microbiological traces,” they suggested using artificial intelligence and a wealth of public data to “monitor human society for signs of unusual activity consistent with the emergence of novel pathogens.” with pandemic potential”.

At the heart of the study is a searchable database of 2.3 billion US Twitter posts that Li’s lab has collected since 2015. For example, enter the word “cough” and the results can be narrowed down by location, time period, and other variables to pinpoint trends.

The hard part is figuring out which tweets make sense, and then training the project’s computer to recognize them.

Some keywords, like “fever,” appear in too many non-health contexts to be relevant, Li says.

So far, the method has discovered “interesting pandemic signals from March 2020,” says Noymer. “Unfortunately, it’s too late to be useful,” as health officials began issuing warnings more than a month ago, he notes. The aim of the grant is to uncover clues from late 2019, before the coronavirus was on anyone’s radar.

“We hope to find the horse while it’s loose in the stable, but before it storms out of the stable,” says Noymer.

To do this, researchers also analyze news reports, anonymous student health and absenteeism statistics, biological data and a range of public information sources — and not just for COVID-19 precursors.

One of the limitations of the study is that the coronavirus originated in China, where Twitter is officially blocked. Therefore, the team will also look for early evidence of monkeypox as a test case.

“If we can’t find any precursors to COVID-19 outbreaks or monkeypox, then our concept is on the up,” says Noymer. “And even if we find them, that doesn’t guarantee they’ll herald the next pandemic. But the potential payoff makes the idea worth exploring.”

By the end of December, Li says, the group hopes to have an AI pilot system ready, a broader collection of data to analyze, and additional results.

Other UCI researchers on the team include Carter Butts, Chancellor Professor of Sociology; Kristin Turney, Dean’s Professor of Sociology; and Dominik Wodarz, Professor of Public Health and Disease Prevention.

If you would like to learn more about supporting this or any other activity at UCI, please visit the Brilliant Future website at Launched publicly on October 4, 2019, the Brilliant Future campaign aims to raise awareness and support for UCI. Through the engagement of 75,000 alumni and raising $2 billion in philanthropic investment, the UCI strives to achieve new excellence in student success, health and wellbeing, research and more. The Donald Bren School of Information & Computer Sciences and the Planned School of Population and Public Health play critical roles in the campaign’s success. Learn more when you visit and