Sign In

Communications of the ACM

ACM TechNews

Study Shows How Easy It Is to Determine Someone's Identity With Cellphone Data


View as: Print Mobile App Share: Send by email Share on reddit Share on StumbleUpon Share on Hacker News Share on Tweeter Share on Facebook
A bored cellphone user.

Researchers at MIT found they could identify individuals using as little as two randomly chosen points using cellphone mobility data.

Credit: Straight North

Massachusetts Institute of Technology (MIT) researchers recently completed a study examining how much information is required to determine an individual's personal identity. After analyzing 15 months of cellphone mobility data from 1.5 million users, the researchers found that only four spatio-temporal points are needed to uniquely identify 95 percent of individuals.

The study could lead to modifications in privacy law in order to keep pace with technological advances.

The researchers also found that knowing just two randomly chosen points can uniquely identify more than 50 percent of individuals. The reference points could come from information that is publicly available, such as the individual's home address, workplace, or Twitter posts.

The researchers found that, contrary to expectations, decreased resolution does not make the data more anonymous, and that just a few more pieces of information are needed to identify individuals. The researchers developed a mathematical formula that tells the probability of uniquely identifying an individual based on the data's temporal and spatial resolution.

"Our formula allows us to estimate privacy, so now the question is how do we use it to balance things out and make it a fair deal for everybody?" says MIT's Yves-Alexandre de Montjoye.

From Phys.Org
View Full Article

 

Abstracts Copyright © 2013 Information Inc., Bethesda, Maryland, USA


 

No entries found