The Family Flu: Can You Find Out If Youll All Get Sick Via Twitter?

Mapping Illness

I don’t know about you, but I am totally sick of being, well sick. Ever since I had my daughter I have had bestowed upon my body almost every cold, flu, or malady that has crossed our paths. Due to this, I have became somewhat of a germaphobe. Washing our hands has become a family pastime and I never go anywhere without a bottle of hand sanitizer. But is that really enough to keep the cooties at bay? A new way of crunching Twitter data may help us germaphobes and those who love them avoid getting sick, or at least know that we may have been exposed (and prepare us to stock up on tissues and cough syrup).

Adam Sadilek and his colleagues at the University of Rochester in New York took a slew of Twitter data, tracked patterns and clusters, and “found that they can predict when an individual person will come down with the flu up to eight days before they show symptoms.”  The researchers drilled down on 4.4 million tweets from about 630,000 Twitter users in the Big Apple and narrowed those down to people who sent out tweets about being ill. The found that they could predict who was about to sick with a stunning 90 percent accuracy.

But the formula does have limitations. First off, not everyone tweets and those that do don’t always accurately tweet their symptoms. Plus, people get sick in ways besides just being in contact with others.

If this program does become a tool for all of us to use, it could have beneficial as well as damaging effects. You may see that your child’s favorite playground is an epicenter of germs and avoid it in favor of one less contaminated. But there is the risk that business areas that are in the midst of a hot spot would be avoided like the plague (pun intended) and that would be bad for business.

Would you welcome a tool like that or are you fine with fate taking its germ-filled course?

Image via New Scientist