University of Virginia Assistant Professor Matthew Gerber has discovered a correlation between Tweets and crime. No, folks aren’t Tweeting that they’re going to knock over a liquor store. Instead, by looking at the GPS coordinates attached to Tweets as well as assessing a heat map of popular locations, police can predict where crime is most likely to happen.
“My initial hypothesis was that there would be no correlation between Twitter use and crime. After all, people don’t share with the world that they intend to or have just committed a crime,” said Gerber. “What they do share are things like social events or outings that could lead to criminal activity.”
Gerber collected 1.5 million tweets in the Chicago area between January and March 2013. He also grabbed crime records over the same period. He divided out the Tweets based on geographic location and looked at the crime data for each area as well as the type of Tweet – sports cheers, restaurant mentions, etc. – to see if Tweet density correlated to crime. It did. The method could accurately predict “19 out of 25 crime types.”
“Some cities that utilize such methods as a basis for resource allocation have seen dramatic decreases in crime,” said Gerber. Unfortunately he cannot yet say why Tweets predict crime on a causal level but it seems to be working and he’s working with police departments in Chicago and New York City to help build out the system.