Twitter. (Credit: EPA/Lex Van Lieshout)
Twitter. (Credit: EPA/Lex Van Lieshout)

What and where you tweet says a lot about you. Researchers at the University of Virginia think Twitter can even help predict crime, according to Agence Presse France.

In a paper published by the journal Decision Support Systems, researchers wrote that combining Twitter and criminal data for a specific location could predict 19 of 25 types of crime.

Criminals aren’t about to tweet “OMG, about to rob this dude,” but geotagged tweets, combined with historical crime data for an area, could alert police to a potentially dangerous situation, according to the researchers.

Advertisement - Continue Reading Below

From the Agence Presse France report:

The results are surprising, especially when one considers that people rarely tweet about crimes directly, said lead researcher Matthew Gerber of the university's Predictive Technology Lab. Gerber said even tweets that have no direct link to crimes may contain information about activities often associated with them.

"What people are tweeting about are their routine activities," Gerber told AFP. "Those routine activities take them into environments where crime is likely to happen.

"So if I tweet about getting drunk tonight, and a lot of people are talking about getting drunk, we know there are certain crimes associated with those things that produce crimes. It's indirect."

The goal isn’t some sort of “Minority Report” pre-crime division, arresting people based on the likelihood they’ll commit a crime. Instead, law enforcement could use data gleaned from Twitter to station officers in potential hot spots, according to the paper:

This research has implications specifically for criminal justice decision makers in charge of resource allocation for crime prevention. More generally, this research has implications for decision makers concerned with geographic spaces occupied by Twitter-using individuals."

(h/t Business Insider)

Scalese can be reached at roberto.scalese@globe.com or via Twitter @BertoScalese.