People who tweet when they’re ill could act as an early warning system for the spread of disease. Adam Sadilek at the University of Rochester in New York analyzed 4.4 million tweets tagged with GPS location data from over 630,000 users in New York City.
By focusing on tweets that appeared to mention illness, he was able to predict when healthy people would fall ill — as confirmed by their subsequent tweets — up to eight days in advance, with about 90 percent accuracy.
Such information could one day power a smartphone app that warns you when you are anywhere with a high incidence of flu. The findings were revealed at the Conference on Artificial Intelligence in Toronto, Canada, in July.
–Â Andy Coghlan
(C) 2012. NEW SCIENTIST MAGAZINE. REED BUSINESS INFORMATION LTD. ALL RIGHTS RESERVED. DISTRIBUTED BY TRIBUNE MEDIA SERVICES, INC.
