A new Google application called Finder, based on artificial intelligence, is said to be capable of working out which restaurants could make you ill. This innovative programme primarily uses research from Internet users and geo-tracking, and initial results appear promising.
Google’s artificial intelligence specialists have teamed up with researchers from Harvard to develop a type of artificial intelligence that can help the food health authorities. Their new tool has been named "Finder", which stands for "Foodborn IllNess DEtector in Real time", as quoted in the trade journal Nature and then reported by Slate.
The programme uses machine learning to identify which restaurants have hygiene standards that could pose a risk to consumers. Tests carried out in the USA in the cities of Las Vegas and Chicago have confirmed the effectiveness of the alerts.
Identifying establishments at risk
Finder functions with millions of items of data collected by Google throughout the world via research carried out on the subject of food poisoning. As soon as a person uses the search engine to look something up on this subject, Google checks to see whether that Internet user has been to a restaurant during the past three days.
If several users have been to the same restaurants and suffered similar symptoms, Finder is then capable of setting off an alert with the food health authorities to ensure that the establishment undergoes a check.
Good levels of success
Tests carried out to date have shown that Finder works reasonably effectively, with 52.3% of the establishments under question turning out to have genuine health-related issues. On the other hand, the rate is only 22.7% when routine checks are carried out. Alerts of this kind are very helpful for the food health services, who can act far more quickly and contain the risks of food poisoning among clients.
"Food-related illnesses are common and costly, with thousands of Americans ending up in the emergency room every year," explains Ashish Jha, one of the researchers behind the programme, which is slated to be tested in other countries before being put to general usage.
Contact Allianz Partners
Sep 27, 2018
By analysing the information gleaned from Google Maps and Street View, as well as data from previous studies, an algorithm is able to determine the obesity rate in a particular geograph [...]
Mar 15, 2018
DeepMind – the Google subsidiary dedicated to artificial intelligence –and the Moorfields Eye Hospital in London have developed a form of IA which can detect anomalies on retinal sc [...]