Twitter has become a go-to news source for what people are talking about around the world, and a new study looks at how helpful the social networking platform could be to law enforcement.
Researchers at Cardiff University in the UK used a machine learning algorithm to analyze data from the London Riots in 2011. From a data set of 1.6 million tweets, computers could automatically scan Twitter and identify potential crimes long before they were reported to police — in some cases more than an hour faster. In fact, the system detected incidents and potential problems faster than police sources in all but two instances. It can also scan Twitter and pick out potential riots, along with information about where people are gathering in real time.
“We will never replace traditional policing resources on the ground,” said the study’s co-author Dr. Pete Burnap, “but we have demonstrated that this research could augment existing intelligence gathering and draw on new technologies to support more stablished policing methods.”