Cell phone data is one of the most basic examples of Big Data capability. Although Big Data consumes all data types, it’s easy to appreciate how just in smartphones vast amounts of data are captured for use – historical moment by moment and in real time. Smartphones have email, pictures, music and oh so many apps that make each phone unique; they also record location, velocity and acceleration.
But out of the almost seven billion cell phones worldwide (2013), five billion are in developing countries. These phones are a completely different animal, with most having little more than call and text capability. They also often serve the prime means of banking.
At the Harvard School of Public Health in Boston, epidemiologist Caroline Buckee uses anonymized cell phone usage data to data mine for patterns in outbreaks of malaria. She has traced the usage from one cell phone tower near a tea plantation employing migrant workers to that of areas outside Lake Victoria, which have a significant malarial concentration. This information assists government and NGO efforts with utilizing limited resources to target control of the disease effectively instead of less effective (and more expensive) blanket efforts.
Along the same lines, cell phone data has helped in the post disaster recovery efforts of Haiti’s earthquake. Information from cell phone usage is highly insightful in countries where less formal data collection exists. Using historical and real time cell phone data, data analysts were able to determine how many people were in Port Au Prince prior to the earthquake, and where they went (or didn’t go) after the quake.
Their work has led to a model for future disaster events and inspired more efforts for utilizing cell phone data usage.
“… last year (2012) Orange, the France-based global telecom giant, released to the world’s research community—subject to certain conditions and restrictions—data based on 2.5 billion anonymized records from five months’ worth of calls made by five million people in Ivory Coast.” – David Talbot, MIT Technical Review, APR 23, 2013
That type and quantity of data is now being used for the Data for Development Challenge (D4D) – an international event to conduct research of cell phone Big Data. The results are presented on an extra day tacked onto MIT’s broader mobile technology conference NetMob – this year in April.