|
|
|
|
|
by curare
2879 days ago
|
|
At a previous employer we used the data referenced in the question. We bought the "anonymised" telecom data from one of the big providers in the USA. We would then analyse that data to figure out where the phone spent 6-8 hours during the night to determine where the phone owner lived. Then we'd pull the USA consensus data(free!) so the system would know what demographic the user probably belongs to. We got a
surprising amount of data from the consensus, we had things like ethnicity, income, age. Then we'd loaded all this in google maps and let users of the app figure out what demographics frequent particular locations on the map. For example a use-case of this would be a coffee company figuring out where to open a new coffee shop. |
|