Catalog describing data formats relevant to facilitating data exchange between many different actors in a decentralised way.
This repository contains various folders containing information and templates for describing data from various sources.
There are two main folders, the data-specific, and the entity-specific.
TODO
An entity is defined as a data-holding organisation, to which a Subject Access Requests can be made. This directory thus contains information about the formats of files received through Subject Access Requests, along with models and example data.
Entities are classified under different categories:
- Gaming: Description.
- Business: Description.
- Educational: Description.
- Lifestyle: Description.
- Entertainment: Description.
- Utility: Description.
- Travel: Description.
- Communication: Description.
- Medical: Description.
- Social: Description.
- Dating: Description.
- Misc: Description.
Of course, new categories can be created at the discretion of the contributors.
Each entity will have each own folder, containing:
- A
README.mdfile that should contain a description of the entity, a description of the scripts, and a description of the templates. - A
inputssub-folder that should contain synthetic data sources, mirroring an export from the entity. - A
scriptssub-folder which should contain the various scripts, notebooks, etc., used to create the templates. - A
modelssub-folder which sould contain the various templates created, with a corresponding extension name. - Whatever useful information (e.g., other models, other descriptions, links, ...).
In particular, a description of how to make a query for its data will usually be present.
For instance, the files Happn/data/all-data/users/uuid.json and Happn/data/all-data/relationships/uuid.json show in a privacy-preserving way how dating app Happn structures information about its users and relationships respectively. It contains no revealing information about the individual who contributed those files.
Those files are generated through a semi-automated process, which is constantly improving.
We are looking for collaborations in the academic and education areas, or even more generally civil society. Feel free to contact us at contact @ hestia.ai.
Initial contributors are the DatingPrivacy , The Eyeballs and other projects supported by HestiaLabs.