"I'm a pretty private person, and I wish others would be more cautious with the types of information they share," lead author Chris Weidemann, a graduate student in USC's Geographic Information Science and Technology (GIST) online master's program, said in a statement. "There are all sorts of information that can be gleaned from things outside of the tweet itself."
While just 6 percent of users opt in to have Twitter broadcast their location with every tweet, other information can also reveal a user's location.
Weidemann developed an application called Twitter2GIS to analyze the metadata collected by Twitter, which includes details about the user's hometown, time zone and language. That data was mapped and analyzed, and Twitter2GIS was able to determine the user's location to an accuracy of street level or better from approximately 20 percent of tweets.
While many Twitter users gave away their location directly through active location monitoring or GPS coordinates, about 4.4 million tweets a day provided so-called "ambient" location data, from which users might not be aware that they're divulging their location.
When he used Twitter2GIS to analyze his own Twitter account, Weidemann was surprised to find that the application was quickly able to determine his location from a hashtag he used about an academic conference. "This research has been fun, and a little scary," Weidemann said.