We recently attended the International Conference on Establishment Statistics. We had an interesting time hearing about some of the challenges statistical organisations, particularly in the US, have with protecting and disseminating establishment data and the move to differentially private methodologies.
We participated in one of the software demonstration sessions and I gave a demo of how our software Cantabular could be used to publish a flexible user interface and API on top of establishment data.
To power our example, we used some open data from Companies House in the UK, but it mirrors how you might achieve a flexible user interface on top of differentially private synthetic data.
We’ve put together two videos covering more-or-less the same material we showed to the attendees.
Using our UI and API to explore and visualise data
The first shows how our user interface and APIs can be used for exploration and visualising of the data:
Protecting data with automated disclosure checks
The second shows how our interactive query tool can be given an additional layer of protection by automatically detecting disclosure risks in tabular outputs using our disclosure rules language:
If establishment data is your thing and you’d like to talk to us about how we might be able to help you protect and publish data, feel free to get in touch.