Geographic location stamps transmitted in tweets can provide enough information for people to deduce where a Twitter user lives and works, say researchers.
The deduction occurs through the clustering of the posting locations. The assemblage provides location patterns that provide a good guess as to where the poster spends most of his or her time.
When that’s coupled with other data, such as the time of day, non-scientists recruited for the study simply picked out the homes and workplaces of the tweeters, said researchers from MIT and Oxford University in a press release.
Where the study becomes interesting is that the tweeter opted in to the location sharing. (It’s off by default.) And, when turned on by the user, although the tweeted location appears to be a broad-brushed label—such as is provided for posts from a city like Los Angeles, where the label is simply "Los Angeles"—there is an option to share an exact, precise location.
That location, logged by Twitter using highly accurate latitude and longitude coordinates, is openly searchable with an API.
+ Also on Network World: How iOS, Android apps share your data without notifications +
The study researchers presented that kind of data to the participants who were able to use it to figure out where people lived and worked. The places were confirmed by the tweeters submitting their actual addresses to the researchers.
“Many people have this idea that only machine-learning techniques can discover interesting patterns in location data,” said Ilaria Liccardi, first author on the paper and a research scientist at MIT’s Internet Policy Research Initiative.
No need for high-tech big data skills
However, regular people without specialized big data skills were able to figure out the locations—particularly when the data was mashed onto a map with the time of day.
“When you send location data as a secondary piece of information, it is extremely simple for people with very little technical knowledge to find out where you work or live,” Liccardi said.
With five days-worth of data, the participants identified Twitter users’ workplaces more than 85 percent of the time.
Interestingly, this public sharing of fine location may well be unintentional. The researchers speculate that it doesn’t occur to tweeters that "the man in the street" could figure out where a Twitter user lives or works. Also, because the tweeter thought the data was benign, the person shared it.
The researchers also found that you don’t need to be a data scientist to figure things out from big bata.
The “survey shows how people can learn sensitive information from seemingly innocuous facts, and, second, people will easily share information they believe is innocuous,” said Latanya Sweeney, professor of government and technology in residence at Harvard, in the press release.
Sweeney was involved in the Personally Identifiable Information ‘behind-the-scenes’-sharing by Android apps study that I wrote about in December.
Tweet locations aren’t the only geo-tagged social network posts that are drawing attention. Last month I wrote about a study that found a person’s “real-world movements” are so unique that people can be distinguished by their patterns. In that study, the scientists said location tracking can be used to differentiate one person from another.
And while we’re on the subject, mobile network operators’ general harvesting of user data (unrelated to Twitter post locations and the aforementioned studies), called telco data as a service, will be a $79 billion business by 2020.
This article is published as part of the IDG Contributor Network. Want to Join?