during testing, the program was able to correctly identify 23 of 25 test mass protests.

a Group of scientists from the U.S. and Ireland have developed an algorithm for detecting suspicious clusters of people based on open data provided by the mobile operators. With its help, the researchers said, it will be possible to create a system of prevention of riots.

Photo: Reuters

Using the data about the calls committed in côte d'ivoire in the period from December 2011 to April 2012 (at this time the country faced a severe political crisis, accompanied by riots), scientists were able to develop an effective method of tracking in cellular networks.

Initially, the algorithm introduced the concept of "normal movement", so that the system has learned to monitor movement of people in the usual daily routes. After that, the authors have added the ability to define custom behavior of subscribers of mobile operators, when many people suddenly starts to gather in any place.

during testing, the program was able to correctly identify 23 of 25 test mass protests.

the study Authors hope that in the future the algorithm will be used in tracking people in real time, able to serve for the purposes of public security or the prevention of mass protests and riots), and for the needs of urban infrastructure, for example to improve transport networks.

See also: drones taught to disperse rallies >>