There have been a slew of projects recently seeking to turn social data into actionable insights and trends. Most of these tend to be on a large scale, looking at stock movements or the spread of disease.
Chris Weidemann from USC Dornsife believes that data from Twitter can deliver micro-predictions as well. He believes that combining our tweets with their geospatial data can reveal our movements with considerable accuracy.
“I’m a pretty private person, and I wish others would be more cautious with the types of information they share,” said Weidemann, a geospatial technology manager for a Virginia company that builds geographic information systems for the federal government.
Weidemann believes that most Twitter users, this one included, are largely unaware of the large amount of location based metadata that is publicly available to others. Some 6% of users opt-in to the geotagging facility on Twitter, therefore broadcasting their location alongside each tweet they make. Given the size of the Twitter member base, that represents around 30 million people giving up their locations on a regular basis.
Depending on how your account is setup, you could be revealing everything from your hometown to your language to potential snoopers.
“There is all sorts of information that can be gleaned from things outside of the tweet itself,” Weidemann said.
All of this has been packaged up into an app called Twitter2GIS. The app harvests tweets based either on a specific geographic location or a specific user. The data collected is then processed by the GIS application Esri ArcGIS to look for trends.
Whilst there are clear commercial possibilities for such analysis, it also opens up some obvious criminal possibilities.
If you would like to view your own geosocial footprint, you can do so via an app Weidemann is developing as part of his masters thesis. Check it out here