Researchers at UCLA created an algorithm that enabled them to predict HIV breakouts based on people’s Twitter activity.
Published in the journal Preventive Medicine, the study collected 550 million tweets between May and December 2012, using an algorithm to pick out words that would suggest drug use or risky behaviors, such as ”sex” or “get high.” These tweets were then plotted on a map and compared to incidences of HIV-related cases. They found a high proportion of these tweets had incidences in areas with reported HIV activity.
Sean Young, Assistant Professor of Family Medicine at the David Geffen School of Medicine at UCLA and co-director of the Center for Digital Behavior at UCLA, said in a press release:
Ultimately, these methods suggest that we can use ‘big data’ from social media for remote monitoring and surveillance of HIV risk behaviors and potential outbreaks. While similar algorithms have been used to predict influenza outbreaks, this is the first 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.
The self-admitted weakness of the study is that the HIV data used was from 2009 and that more frequently updated data would be needed to make proper predictions. It shows the possibility of real-time predictions of possible HIV outbreaks, leading to preventative measures being taken.
Article source: http://www.psfk.com/2014/03/twitter-hiv-prediction.html