August 20, 2012
Nokia Mobile Data Challenge May Predict Where Mobile Users Will Go
Enid Burns for redOrbit.com - Your Universe Online
Big brother is getting a few new tools that just might be able to determine where you'll be tomorrow. A team of computer scientists from the University of Birmingham in the U.K. won Nokia's Mobile Data Challenge with a program that uses an algorithm that can predict where you'll be in the next 24 hours.
Location prediction is possible through "the study of the interdependence of human movement and social ties of individuals," it says in the paper, which calls it "one of the most interesting research areas in computational science." Researchers Manilo De Domenico, Antonio Lima and Mirco Musolesi, all from the School of Computer Science at the University of Birmingham in the United Kingdom, contributed to the paper "Interdependence and Predictability of Human Mobility and Social Interactions," which won Nokia's Mobile Data Challenge for 2012.
The paper outlines an algorithm that uses data from a user's phone to predict where that person will be in the next 24 hours. Where this method differs from location prediction methods discussed in the past few months is that the new algorithm uses data from friends and contacts in the user's phone to make its predictions. This new algorithm looks for patterns that might diverge from routine, where predictive systems previously only were able to determine locations based on patterns and routine.
The algorithm works to determine a target user's location to within 20 meters, about the space of a city block.
Advertisers may see the program as a boon. The ability to send advertisements, coupon and other messaging to locations a user plans to go, or can be swayed to go with the right offer, is highly sought after for advertisers. There are many ways the technology could be put to use for good.
There are just as many ways that the data mining technology could be used for bad, or perceived as bad by those who fear big brother. Many people will find the method too invasive, and fear it could be put to use by the government to track and predict where users go. A Wired UK article suggests the new technology will have people resort to "tin-foil-hat scenarios" of government location tracking.
Predictive location tracking requires a large degree of actual location tracking, which brings up privacy concerns. The Wired article says that while it might not be OK for some, it may be alright for others. "Mulolesi points out that some people are happy to share their location publicly," it said in the article.
This isn't in use just yet. Users don't have to worry about how to shield their phones or opt out of the program at this time. Even for users who would opt in to such a program, it can raise the hairs on the back of someone's neck to get an ad for a restaurant or store they are planning to visit in the next 24-hours, especially if it's a location they don't normally visit.