London, Jan 26 (ANI): Scientists have developed a new system for predicting a Twitter user's location by looking at where their friends are.
The system, developed by Adam Sadilek from the University of Rochester in New York, can correctly place a user within a 100-metre radius with up to 85 percent accuracy.
"You can actually infer a lot of things about people, even though they are pretty careful about how they manage their online behaviour," New Scientists quoted Sadilek as saying.
Sadilek and his colleagues turn their target's social network into a predictive model called a dynamic Bayesian network.
At each point in time, the nodes in the target person's network consist of their friends' locations, day of the week and the time.
The information from these nodes determines the target user's most likely location. Sadilek can also feed in any existing information about the person's whereabouts to help improve the model's accuracy.
The model was tested on over 4 million tweets from users in Los Angeles and New York City, who had location data enabled.
The researchers found that when a couple of weeks of location data on an individual was combined with location data from their two most sharing friends, it was enough to place that person within a 100-metre radius with 77 percent accuracy.
The accuracy rises to nearly 85 percent when the already existing information is combined from nine friends. Even someone who has never shared their location can be pinpointed with 47 percent accuracy from information available from two friends, which rises to 57 percent with nine.
Once the model has a good idea of where some people are, it can use this data to predict who their friends are, and even use that social network to pinpoint the whereabouts of more people.
"You can imagine looping this process over and over," Sadilek added.
The work will be presented at the Web Search and Data Mining conference in Seattle next month. (ANI)
Read More: Data | Nagpur City Ho | Kamthi City So | Adam | Gondia City S.o. | Chandrapur City So | Malkapur City | Akola City | Achalpur City | Burhanpur City | Indore City-2 | Hoshangabad City | Damoh City | Chhatarpur City | Guna City | Gwalior City | Lashkar City | Chhindwara City | Balaghat City (tso) | Jabalpur City Mdg