WE INGEST MULTIPLE DATA SOURCES
TO PRODUCE THE WORLDS FIRST SUB
NATIONAL VIEW OF GDP
To produce GlobalGDP.ai we used multiple sources of data including satellite data (road length, agricultural activity, heat generation, particulate count), official data, and data from a wide range of sources including telecommunications, payments and energy systems data. These data sets are then combined to show an alternative view for monitoring global GDP. Credit for some of these data sources go to Earth Observation Group, Payne Institute for Public Policy. Global GDP was built for educational purposes only and for providing an alternative view to calculating GDP.
AI & big data
It would be near impossible for a human to look to process the amount of data and information found within GlobalGDP.ai. From analysing millions of images on a regular cadence to abstract the information contained within each pixel on a global scale through to aligning location-based data to drive location intelligence data requires a massive amount of computing. This is where the artificial intelligence and big data bridge the gap. GlobalGDP.ai utilises a location intelligence AI engine that can ingest and digest trillions of data point and pixels at scale. Our algorithms turn each data point and pixel into valuable information on the state of GDP for that area.
From crunching the vast amounts of raw sensor data we then combine this into multiple different data models. Each data model has its own predefined role and task. Once that task is complete, the model is merged into a hybrid data model that produces the final GDP figure.
Once the models have produced their outcome, this data is combined with one final model that takes the GDP data output from the World Bank and breaks it down by each region as set out by the GADM.org portal.