With such big advancements in available technology, it’s becoming easier to measure how people and vehicles move around the world. But, what exactly is the data you’re gathering and what should you do with it?
Here at StreetLight, we define Massive Mobile Data Analytics as using data derived from mobile devices (smart phones, dumb phones, connected cars, wearables, etc.) with the following characteristics:
Massive (at least 1%, preferably >10% of the population covered. This means > 30M devices for the US).
Mobile (derived from a mix of smart phones, dumb phones, connected cars, wearables).
In order to get access to data with the previous two characteristics, the data must also be anonymous (no Personally Identifiable Information), archival (at least 15 minutes old, up to a few years old), and delivered as an aggregate metric describing a group of people.
There are currently two main flavors of Massive Mobile Data Analytics: “Cellular Network” and “GPS Network”. Sometimes, the best approach is combining both sources.
Figure 1 – Two Major Classes of Mobile Data, Visualization of Representative Raw Data. Purple circles are data derived from cellular towers and blue dots are data derived from GPS satellite networks. To protect privacy, the data in this image is synthetic (representative of but not taken directly from raw data).
Table 1 – Two Major Classes of Mobile Data, Characteristics and Use Cases.