We’ve talked about some of the safety measures that can be taken to prevent food-borne illness. Today, we’re going to be taking a look at this from another angle. What if there are restaurants out there who have been spreading food-borne illness and it doesn’t get reported?
Most of the time food-borne illness doesn’t get reported. It’s a rare case that salmonella will be cause to hospitalize – most people simply take it as the price of playing the game and don’t bother going to a doctor about it. Even though it doesn’t become a ‘newsworthy’ event, it should still be of concern to health inspectors.
Let’s bring in big data and machine learning. What if there could be a system that pinpoints restaurants for health inspectors to take a look at? The Association for the Advancement of Artificial Intelligence (AAAI) has brought something new to the playing field.
Now, there’s a big chance that people won’t talk about their illness to an epidemiologist or a doctor. They will, however, talk about being sick and where they ate on Twitter. Scientists at AAAI developed an application where they could scan all of the tweets in a localized area and track them back to potential food poisoning outbreaks.
These tweets were pretty useful, too. Using this system (nEmesis), they were able to identify more restaurants who had health violations, keeping people safe. “nEmesis has proved to be a useful tool for quickly and accurately identifying facilities in need of support, education, or regulation by the health department.”
This approach could potentially be applied to all kinds of health problems, and it could be eventually used to help those who might want to market their restaurants. Imagine, if you will, a system which automatically sent a tweet to someone in the area who said that they wanted a specific type of food. That kind of timely information could definitely drive sales, and keep those charbroilers running.
