Drug dealers advertising their products on social media are being rounded up by a computer program developed by Southern California researchers, in an effort to combat illegal drugs being sold through the internet.
Researchers at the University of California San Diego developed a machine learning technology that mined Twitter to identify users peddling the illegal sale of opioids online. The study was published in the October issue of the American Journal of Public Health.
Researchers mined tweets between June and November 2015 and found some 619,937 tweets that contained one of a number of keywords of prescription opioids. These included codeine, Percocet, fentanyl, Vicodin, Oxycontin, oxycodone and hydrocodone.
Of those, 1,778 posts were marketing the sale of the controlled substances and 90 percent included hyperlinks for online purchase. A majority of retailers had web addresses based in Pakistan, the researchers found.
“Our study demonstrates the utility of a technology to aid in these efforts that searches social media for behavior that poses a public threat, such as the illegal sale of controlled substances,” Tim K. Mackey, UC San Diego School of Medicine associate professor of anesthesiology, said in a press release.
He added, “The online sale of controlled substances is directly prohibited by federal law. However, social media appears to act as a conduit for increased risk to substance abuse behavior.”
The technology consisted of at least three elements to filter out offending tweets. First the researches employed an algorithm to separate out tweets that continued keywords associated with prescription opioids, then a machine learning program – used unsupervised – further filtered the tweets to highlight those that were related to the marketing of opioids. A web forensic examination analyzed posts that had hyperlinks to external websites, the researchers wrote.
“Our methodology can identify illegal online sale of prescription opioids from large volumes of tweets. Our results indicate that controlled substances are trafficked online via different strategies and vendors,” they wrote in their conclusion.
The researchers hope the technology can be used to live surveillance and detection of illegal online activities and to ensure compliance with the Ryan Haight Online Pharmacy Consumer Protection Act, which prohibits the online sale and marketing of controlled substances.
The 2008 law is named for Ryan Haight of San Diego, who died from an overdose in 2001 mixing prescription pills he had bought online.