Home > Latest News > Environmental Health > Food >
Twitter Tracker Can Predict Food Poisoning Cases
Researchers at the University of Rochester in the USA think they can predict likely cases of food poisoning using the social networking tool, Twitter. The newly developed nEmesis system tracks millions of tweets looking for mentions of illness after restaurant visits, it then ranks those mentioned according to the likelihood of food poisoning occurring. Over a four-month period, the system collected 3.8 million tweets from more than 94,000 unique users in New York City, traced 23,000 restaurant visitors, and found 480 reports of likely food poisoning. When compared to results of inspections by the local health department there was a good correlation as Henry Kautz, chair of the computer science department at the University of Rochester explained; “the Twitter reports are not an exact indicator, any individual case could well be due to factors unrelated to the restaurant meal, but in aggregate the numbers are revealing. “
In all, 120 restaurants that appeared to be responsible for cases of food poisoning were found during the trial. The system only captures information from Twitter users and is intended for use alongside more conventional methods to provide a guide as to where problems might occur. Developers are also concerned that the technology could be open to abuse if the system ever goes public. As nEmesis principal inventer Adam Sadilek put it “You enter another stage where people know about the system. People will start tweeting that they were ill when they got near certain restaurants.” Sadilek, who now works for Google will present nEmesis at the Conference on Human Computation & Crowdsourcing in Palm Springs, California, in November.