This article describes an Israeli study which tried to find the existence of emergencies based on spikes in telephone and text traffic. The idea goes like this:
- Identify high traffic nodes, essentially people who make lots of calls and text messaging (the data was all anonymized to avoid identifying actual people);
- Take non-random samples of call and text traffic at these high traffic nodes;
- Spikes in traffic, amplified by a fancy algorithm, suggest the presence of a big event.
To quote the article:
The researchers applied an algorithm they call the Social Amplifier to this nonrandom sample to tease out where and when emergency events could be found. Those uncovered included eight real emergencies—three storms, a bombing, an earthquake, a blackout, and two airplane-related events—as well as eight other big local happenings, such as concerts and festivals, that were not emergencies.
Can such an approach be done here in the US? We wonder if the public would believe that the use of cell phone data – especially data based on identifying heavy callers – was really anonymous. And since the approach appears to identify many events with big social impact – not just emergencies – it’s really only valuable as a supplemental data source.
Still, in a world that is being rapidly shaped by the use of “big data”, it’s an interesting and potentially useful idea.
