Thursday, May 2, 2013

Using Cell-Phone Data to Improve our Cities

Imagine what questions you could answer if you had access to movement data collected from millions of cell-phone users. These kind of data bring and endless potential for research in several areas such as transportation, epidemiology, social networks and civil defense. MIT is holding this week the NetMob 2013 Conference, a conference focused on mobile phone data analysis. There are several interesting papers that are really pushing the envelope.

A paper by Francesco Calabrese and colleagues stands out as they have developed an interactive system to optimize public transit networks using mobile phone data under the Orange D4D Challenge*. Still, I believe there's something to improve in using such kind of data in a way that differentiates people from higher and lower income groups.

*Orange, a big telecom carrier, lauched a challenge last year called "Data for Development" (D4D). In short, you could send them a 250-words research project and get access within a week to a really large mobile phone dataset containing 2.5 billion records (calls and text messages) exchanged between 5 million anonymous users in Ivory Cost. Many papers presented at NetMob 2013 Conference were developed using Orange D4D data.

By the way, the Senseable City Lab (MIT) is also working on a related project (Signature of Humanity) trying to look at mobility patterns comparing New York, London, and Hong Kong. Here's a promotional video: