Latest News

6/recent/ticker-posts

Ad Code

Facebook Movement Data Could Help Find New Covid-19 Locations, Study Finds

Facebook Movement Data Could Help Find New Covid-19 Locations, Study Finds

 Facebook's anonymized data on people's travels could be used to identify the spread of Covid-19 in places where health officials don't yet know, according to a new Australian study.

Published in the Journal of the Royal Society Interface on Wednesday, researchers at the University of Melbourne analyzed anonymized population mobility data provided by Facebook as part of its Data for Good program to determine if it could be a useful predictor in determining the spread of outbreaks of Covid based on where people were traveling.

The investigation looked at three outbreaks in Australia: the Cedar Meats outbreak in West Melbourne, the second largest wave in Victoria, and the Crossroads Hotel outbreak in New South Wales.

The research found places where people had periodic and predictable movements, such as traveling to and from work, provided more useful indicators of the spread of the virus than social settings. In the case studies, the data was therefore more useful in predicting the spread of the virus in the Cedar Meats outbreak than in the Crossroads Hotel outbreak.

When it came to analyzing Victoria's second wave, which began with the confined suburban blockades in late June and early July, the analysis found that mobility data may have alerted the government that the spread had already moved further. beyond the suburbs initially confined to the blockade.

“Our examination of the second wave of community transmission in Victoria showed that several weeks before it was recognized, the spatial distribution of a small number of active cases was indicative of the distribution of the outbreak more than 30 days later, when the interventions were introduced. ”, The researchers said in the newspaper.

“This observation indicates that even when the number of cases was small, it is possible that low-level community transmission has already occurred throughout the Melbourne metropolitan region. This suggests that previous selective blocking measures, which extend beyond the borders of the regions in which cases have been identified, may have been more effective in containing transmission. "

University of Melbourne principal investigator Cameron Zachreson told Guardian Australia that it was too difficult to say whether the data could have changed the Victorian government's decision-making on when to block Melbourne during the second wave.

“We analyzed the mobility data and compared it to the zip codes that had been blocked, and looking at that, it became very clear that blocking these particular zip codes probably wouldn't have the desired effect, but at the same time that blocking by zip code. it didn't last long, ”he said.

“That spread to the Melbourne region in a couple of days. I think [it was] very clear that such an approach would not be enough. "

Zachreson said the data would be useful in cases where not much is known about an outbreak.

“Seeing mobility information like this can give you a decent idea. There is definitely a sign there. It won't give you everything you want.

“There will be places that turned out to be high risk that are not in the data, like in the Crossroads outbreak, there were cases in the Blue Mountains. That was because those people traveled a long way on a journey that they normally wouldn't. And that's the kind of thing that the aggregate mobility patterns that we have don't capture well. "

Zachreson said governments could also use the data to determine where to declare potential hotspots, rather than focusing on arbitrary areas of local government or entire cities, like Sydney currently.

"It can give you a less arbitrary determination of where the high-risk areas are," he said.

Zachreson stressed that the data would not allow researchers to identify someone, as Facebook would have already anonymized him. Governments would also not be able to access the raw researchers' data, which means they would not be able to identify people through the data set either.

Post a Comment

0 Comments