DASH: Privacy Preserving Data Sharing

Do you want to get more out of your data by enriching it with data obtained from external parties? Or do you see an opportunity in trading your own data? In this project, we will find out how this can be done in a privacy-friendly way.

Within this project, the project results are demonstrated in two domains: the healthcare sector and the logistics sector. The aim of this project is to support Flemish SMEs in setting up collaborations that involve the sharing and processing of sensitive data.

DASH project Privacy Preserving Data Sharing

Dec '21 / Dec '23

Flemish government - Flanders Innovation & Entrepreneurship

This project Privacy Preserving Data sharing (DASH) of Odisee and KULeuven is aimed at:

enterprises
non-profit organisations
governments

... from the care sector and the logistics sector.

Some actors explore opportunities to share data. Others want to use external data to optimise their own activities. Still others, for example technology companies, provide algorithms and/or infrastructure to support distributed collaborations.

The aim of this project is to support Flemish SMEs in setting up collaborations that involve the sharing and processing of sensitive data. This includes the selection and integration of:

Qualitative anonymisation strategies
Valuable processing algorithms
Technologies for building a distributed infrastructure

A set of best practices is extracted and compiled in an interactive Data Sharing Guide that helps companies to

selecting appropriate anonymisation techniques
selecting valid data processing techniques
guidelines on the choice of infrastructure the use and sharing of third-party data
workshops with demonstrations of use cases in healthcare and logistics
hands-on sessions on anonymisation tools and distributed learning techniques such as Federated learning.

01/12/2021: start date
01/12/2022: demo practical experiments on public datasets
01/12/2023: data sharing Guide and demo of cases from care and logistics
01/12/2023: end date

This project initially experiments on available public datasets. In a later stage of the project, concrete use cases will be worked out, supplied by and tailored to companies in the user group. Prototypes will be built and presented to test and demonstrate the possibilities and applicability of these technologies to the user group. The user group will be composed of companies that want to exploit data themselves, companies that want to rely on third-party data or companies that want to facilitate this. Eventually, a set of best practices will be extracted that helps companies to select both suitable anonymisation techniques and valid processing techniques, as well as guidelines on the use and sharing of third-party data.