Social media is at it again! A few months ago Google was
able to predict flu outbreaks and now mashable reports that Twitter can predict
HIV outbreaks! Where does our privacy fit into these data collections?
UCLA researchers have harnessed the power of social media
for detection and prevention efforts of HIV. Their work “suggests a link between geographic
outbreaks in the U.S. and tweets with phrases that indicate drug-related and
sexually risky behavior.” Words such as “sex” and “get high” were some of the
phrases that the researchers’ algorithm used in order to predict outbreaks. They
collected more than 550 million tweets that they ran through the algorithm and they
then ran statistical models to assure if the locations matched places where HIV
was reported.
There was a significant relationship between the tweets and
locations, but they used outdated data from 2009 which is obviously a weakness
in their study. The researches would require frequently updated HIV data in
order to predict future outbreaks.
Sean Young, the co-director of the Center for Digital
Behavior, stated, "This is the first [study] to suggest that Twitter can
be used to predict people's health-related behaviors and as a method for
monitoring HIV risk behaviors and drug use."
