Geolocation
Geolocation is a category of data science algorithms in GO that operate on location data to measure activity, analyze movements, and understand behavior in aggregate.
What is location data?
Location data is captured from connected IoT (Internet of things) devices, such as mobile phones, ships, cars, etc. The raw data analyzed by GO generally contains a device identifier, latitude & longitude coordinates of the device position, and date/times at which the device is reported to be at that location.
Raw location data is very large in size - petabyte-scale! In addition, there are many complications in working with raw data, such as panel shifts (where the number of devices in the dataset fluctuates over time).
The following table summarizes the generally available geolocation algorithms:
| Algorithm | Data Source | Applicability | Output |
|---|---|---|---|
| Foot Traffic | Mobile Geolocation (Global Provider I & II, US Provider) | Global (ex-China) | Time series (unique device counts) |
| Traceability | Mobile Geolocation (Global Provider I & II) | Global (ex-China) | Geographic locations (geometries) |
- Data Source: click here for more details on geolocation data sources
- Applicability: the areas and conditions under which the algorithms are expected to perform well
- Output: what the algorithm returns - see the individual algorithm sections for more information
Global applicability
Our geolocation algorithms are designed & built to work globally. Nonetheless, they are limited by whether there is data coverage in various geographies.
More detail on data coverage by geography is available here and on request.
Updated almost 2 years ago
Learn more about the available geolocation algorithms