According tomedia CNET reported that due to limited detection and data, the new coronavirus use of transmission routes are still relatively hidden, difficult to track. Now, researchers are digging into Google search data to gain insight into the spread of the new coronavirus and the respiratory disease COVID-19 it causes, as well as its symptoms and other effects.
One data scientist said he had found evidence of a new symptom hidden in numerous search queries, while another team was working on models that could reveal the true extent of mass transmission.
Researchers from Microsoft, Harvard Medical School, Public Health England and University College London hope to use machine learning and Google search data to track the spread of COVID-19.
“I think it’s a longer-term project, at least until the first and last pandemic is over, and I think it’s a long-term project that will continue.” Vasileios Lampos, from University College London, who led the project, said.
Lampos said the project is still in progress — regularly updated on GitHub — and more data is needed to validate the team’s observations, but hopes that the machine learning model that analyzes search data will be able to predict the prevalence of COVID-19 in specific countries or other populations. The basic concept is that when COVID-19 is transmitted in a group of people, infected people start searching Google for the symptoms they see. Of course, people who are not infected with the virus may also search for symptoms after hearing about symptoms in news reports, so Lampos and his colleagues are developing models designed to control this and other complications.
Previous studies have shown that flu-related Google searches track the actual infection rate of the U.S. flu. Lampos’ project builds on similar efforts to estimate the existence of influenza-like diseases.
While search data may serve as an imperfect but helpful alternative to ubiquitous detection to track viruses, it can also help us learn more about viruses and how they affect us.
Economist and data scientist Seth Stephens-Davidowitz reported his own provisional findings in the New York Times, which appeared to find a new type of COVID-19 symptom in Google searches in people with higher rates of diagnosis.
“Over the past week, ‘my eyes are sore’ and searches have been the highest in New York, New Jersey, Connecticut, Louisiana and Michigan,” he wrote. In fact, the American Academy of Ophthalmology is now advising ophthalmologists, “the report shows that the virus can cause mild follicles conjunctivitis.” “
Search data can also provide other valuable trend information before official statistics such as unemployment. Financial giant UBS has tracked Google’s search for unemployment in all states and across the country.
Searching for viruses isn’t a perfect science, but at this point, public health leaders are eager for any data they can get. Lampos says his team sends the information they have every week to Public Health England.