This technical giant's algorithm can identify shortcomings in the food safety of real-time plants
Google is developing a new algorithm with which it will be able to warn about restaurants where food poisoning is more likely to suffer.
This algorithm can identify errors in the food security of real-time plants by combining data on poison and site history poisons.
To promote this project, Google has collaborated with researchers at Harvard University. Specifically, they have tested a model in the US cities of Chicago and Las Vegas to identify certain searches made by users, related to problems such as "stomach cramps" or "diarrhea".
In the conducted test, the system crossed this information with the data stored in the same mobile device on site history, with particular attention to restaurants and facilities related to the food the user has recently visited.
After that, Google sent health inspectors to several of the restaurants, including some of whom the algorithm pointed out as potentially insecure, while the rest was chosen by traditional methods (for example, based on consumer complaints).
The result was surprising. Some 52.3 percent of restaurants confirmed as unsafe were detected by the Google model, while the total discovery of unsafe restaurants through routine surveys in the two cities was 22.7 percent.
As explained by study writer and Google researcher Evgeniy Gabrilovich, this model can play an important role in combating food-borne diseases.
"In this study we only stick to the scope of what is possible in machine-taught epidemiology," explained Gabrilovich, who compared this test with the work of the father of modern epidemiology, Dr. John Snow, performed in 1854, when he had to go door to door in central London, ask people where they got the water to find the source of a cholera outbreak.
"Today, we can use online data to make epidemiological observations almost in real time, with the potential to significantly improve public health in a fast and cost-effective way," he concluded.
Find out about the latest on digital economy, start, fintech, business news and blockchain. CLICK HERE