University of Rochester researchers say their system, nEmesis, by "listening" to the tweets from other restaurant patrons, can help people make more informed decisions and warn of the possibility of problems at a particular restaurant.
It could also complement traditional public health methods for monitoring food safety, such as restaurant inspections, a university release said Wednesday.
The system, combining machine-learning and crowdsourcing techniques to analyze millions of tweets, can find people reporting food poisoning symptoms following a restaurant visit, Rochester computer scientist Henry Kautz said.
"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," he said.
The system detects restaurant visits by matching up where a person tweets from on a GPS-enabled mobile device and the known locations of restaurants.
In a test in 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, the researchers said.