We all know that companies like Google and Apple can track our locations 24-7 using our cell phones, but what if they could predict where we'd be in 24 hours? A group of people have created an algorithm that allows your phone to predict where you'll be in 24 hours from current time. Look just past the fold for more details...
The average error is with in 20 meters. Past studies have been able to predict people’s future movements with only mild success. Most humans follow consistent patterns, but most prediction algortihims haven’t been able to adjust for breaks in routine.
Here’s how the researchers at the University of Birmingham got around that, according to Slate:
“The researchers solved that problem by combining tracking data from individual participants’ phones with tracking data from their friends—i.e., other people in their mobile phonebooks. By looking at how an individual’s movements correlate with those of people they know, the team’s algorithm is able to guess when she might be headed, say, downtown for a show on a Sunday afternoon rather than staying uptown for lunch as usual.”
For their efforts, the research team of Mirco Musolesi, Manlio Domenico, and Antonio Lima won this year’s Nokia Mobile Data Challenge. While the algorithm was only used on test devices, many have asked how could such an algorithm be used in the real world?
Slate suggests it would be great for a company like Google for serving ads that predict where a customer might be going tomorrow or purchase. It could work really well with Google+ based on users’ location check-ins, but if left in the wrong hands the algorithm could cause a whole new mess of privacy concerns.
No comments:
Post a Comment