i-sense members, Dr Vasileios Lampos and Prof Ingemar Cox at UCL Computer Science, have received USD$200,000 in funding to support their research into modelling the prevalence and understanding the broader impact of COVID-19 using web search data.
The funding, from Google.org, is part of the $8.5m given to 31 organisations around the world that are using machine learning and data analytics to help respond to the current pandemic.
Modelling COVID-19 spread and its impact on vulnerable populations
The funded projects span four themes, for which our team’s research falls under the theme ‘monitoring and forecasting disease spread.’ Understanding disease prevalence within a community can inform public health response and reduce transmission.
The team’s work will focus on modelling the prevalence of COVID-19 using web searches, including the development of forecasting and anomaly detection methods at national and subnational level. In addition, research will be conducted to understand how vulnerable populations and low- and middle-income countries might have been affected by the pandemic.
The main data resource to support these efforts will be web search data, including a novel data set that provides privacy-preserving Google search statistics down to the county level.
“This great Google.org initiative will help us expand on our current work on modelling COVID-19 as well as look at how the pandemic might have affected non-patients,” says Dr Vasileios Lampos.
“We are also fascinated by the availability of Google search data that can support our efforts.”
- Google.org blog: Google supports COVID-19 AI and data analytics projects
- Google Health blog: Using symptoms search trends to inform COVID-19 research
- i-sense news: Tracking COVID-19 using online search data
- Preprint on arXiv: “Tracking COVID-19 using online search”
- Preprint on arXiv: “Providing early indication of regional anomalies in COVID19 case counts in England using search engine queries”