Sign In

Communications of the ACM

ACM TechNews

Taking the Census, With Cellphones

A population density map of France derived from more than 1 billion cellphone call records.

A population density map of France derived from more than 1 billion cellphone call records shows that people congregate in urban areas during working periods (indicated by orange spikes), and head for coastlines during holidays (indicated by the blue spik

Credit: Catherine Linard

A new study by Belgian researchers, published in the Proceedings of the National Academy of Sciences, demonstrates that big data analytics can turn cellphone records into highly granular population density data.

Geographers and data scientists from the Université Libre de Bruxelles and Université Catholique de Louvain obtained aggregate, anonymized call records from major cell carriers in France and Portugal. The data, containing records on more than a billion calls, included information such as the originating and receiving phone towers, call length, and user identifier. The researchers used the call records to develop a model for estimating population density around every cellphone tower, accounting for variations in phone usage in high- versus low-coverage areas.

The model revealed several clear, if not terribly surprising, trends in population dynamics: holidays saw the cities largely empty out, with tourist destinations such as Disneyland Paris and coastlines seeing a population boom. During weekdays people concentrated in cities for work, but many returned to rural areas on weekends.

The researchers found the method was comparable to remote sensing technologies that use satellite imagery to estimate population density. However, the cellphone method was able to provide information on more granular time scales, down to the hour.

From Science
View Full Article


Abstracts Copyright © 2014 Information Inc., Bethesda, Maryland, USA


No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account