This is demonstrated by the study carried out by the University of California at Los Angeles (UCLA) and which has been published in the journal Preventive Medicine . The work suggests that thanks to social media such as Twitter, it would be possible to predict the risk of contagion of diseases such as HIV and drug-related behavior, in order to prevent new outbreaks of AIDS .
Having already associated the use of some drugs with risky sexual behaviors and other infectious diseases, the UCLA researchers thought that using Twitter’s Big Data could help prevent the transmission of these diseases . The key lies in an algorithm specially designed to monitor tweets in real time, assigning where messages come from and linking them with existing data on the geographic location of HIV cases (thanks to the interactive map on AIDSVu.org).
For the development of the study, the researchers collected more than 550 million tweets , registered between May 26 and December 9, 2012, and generated an algorithm whose object was to find words or phrases that suggest drug use or behaviors that potentially dangerous like taking drugs or having sex. Once this was done, the results were plotted on a map that provided data from where these tweets originated to compare if they coincided with the areas where HIV cases had already been reported. The result was positive. The algorithm was able to intercept 8,538 tweets with risky sexual behavior and 1,342 with suggestion of stimulant drug use , which led to the establishment of a significant relationship between the data of the tweets and the cities in which more cases of HIV
Thanks to this experiment, the researchers conclude that it is possible to use real-time data from media such as Twitter to create a map that helps predict future cases of HIV or drug use , in order to prevent them. Get there, if possible, before the virus.
Although previous studies have already examined how to use Twitter as a prediction tool for infectious outbreaks, this is the first time that it has been claimed that the popular microblogging network can be used to predict health-related behaviors of the inhabitants of specific locations (towns, cities …), and as a method to control the risk of contagion from the consumption of certain drugs.
