DATA SHARING: THE KEY TO FURTHER DIGITISATION

Eliminating obstacles to data sharing

Data sharing enables chains to work more efficiently and participating parties to save a lot of money. At the same time, quality, transparency and privacy are essential to data sharing. We therefore focus on the elimination of obstacles in order to enable data sharing between organisations. The starting point is that organisations and individuals have a right to their own data and control over these.

The parties involved can find out exactly who has processed and provided which data and when, using which algorithms and for what purpose. TNO uses innovative techniques to monitor and influence the quality, traceability and repeatability of data services.

In addition to optimising processes within chains, TNO sees great opportunities in an approach which transcends them. An integrated approach to data sharing such as this could accelerate innovations in and between existing sectors and deliver new, hybrid solutions.

HOW DO WE GO ABOUT IT?

How can we overcome the barriers to data sharing? TNO does this by collaborating within consortia, by showing that innovative methods work (proof of concept) and by deploying our knowledge in specific (technical) areas such as Secure Multi-Party Computation (MPC), International Data Spaces (IDS), reliable Artificial Intelligence (AI) and Self-Sovereign Identity (SSI). We show that secure data sharing can work, connecting parties with seemingly different interests and working out ideas together.

Controlled access to available data

As a company, you only wish to share the part of your data that you actually want to share, and only to parties that you want to give access to. The aim is therefore controlled access to available data; in other words, a data sharing infrastructure based on standards (both legal and technical).

Only certified users will have access to a data ecosystem in which they can attach usage restrictions to their data before they are made available to other users. We help companies and public authorities to share and manage their data and make better decisions responsibly and transparently from both a social and a commercial perspective.

This is a solution for many sectors, such as healthcare, high-tech industry, agrifood, ICT and telecommunication. We also contribute to International Data Spaces (IDS), an initiative for the development of a European standard for independent and controlled data sharing. After all, there can be no secure data sharing without agreements. IDS is seen as a promising basis for companies to work on AI, IoT and big data.

Data interoperability

The more we automate, the more important it becomes that systems understand one another. This requires a secure data sharing infrastructure based on standards. TNO is working with companies on the development of a new generation of data standards in order to realise decision support systems and make chains more transparent.

New generation standards can be developed using semantic technology. The sector organisation SETU, for example, has developed exchange standards – which have a predetermined structure – for the temporary employment sector. Thanks to these standards, the integration of IT systems of staffing organisations and customers has become faster and more straightforward.

Alongside a broad consortium of temporary employment agencies and software suppliers, SETU and TNO are now working on the development of exchange standards for the flex market. The HortiCube is an example of a decision support system which we have developed. The HortiCube – part of the BIGt&u project that promotes big data in the horticultural sector – collects data from various sources and supplies them in a standardised form to app developers, who can use this to develop applications for market information more efficiently.

Reliable analysis of data

AI and complex algorithms can provide both insights and data and can make decisions automatically – but before that, we have to be able to rely on data and algorithms. Our goal is therefore to improve the reliability of data products and services by stabilising and increasing the quality and accountability of digitally-produced data products.

We offer insights and tools which help companies to increase the reliability of data analysis by improving the quality, transparency and reproducibility of data products. Within the Smart Dairy Farming project, for example, we work on data quality control from two angles. We determine whether data are of a sufficient quality to be processed and whether the calculation model still applies to the current data.

In addition, we are building a generic IT platform to reproduce and expand NAM risk calculations. This will allow us to calculate how various gas extraction scenarios will affect the land and buildings in the area around the Groningen gas field. This requires highly advanced software management and the ability to keep track of the results produced.

Sensemaking of sensitive data

In addition, we are working on data solutions in which different parties bring together their data in one platform without giving each other (privacy-)sensitive information. We use innovative and secure techniques such as Secure Multi-Party Computation (MPC), which enables different parties to jointly calculate data without actually being able to view one another’s data.

In other words, MPC does not provide insights into data but simply combined conclusions. This paves the way for new products and services that fully comply with the strict privacy legislation. Examples include achieving better treatment methods for greater results in the healthcare sector without violating privacy and financial institutions which can detect fraud and money laundering activities.

Digital validation of information

The digital validation of information makes it possible for everyone to be in control of his or her own data. TNO is carrying out various projects related to SSI technology. SSI offers citizens new possibilities for managing privacy and no longer having to log in with a password. Organisations can help to achieve enormous savings and high customer satisfaction by speeding up and automating transactions with customers, suppliers and partners.

We are investigating how SSI can be made ready for application. Together with other parties from healthcare, public authorities and financial institutions, we are developing use cases into proofs of concept. For example, we are contributing our knowledge to the uNlock consortium, which is developing an open-source, non-profit application which will make it possible to issue a proof of corona test/vaccination with 100% certainty. This application protects the privacy of the users.

Read more

• Download the Dutch report ‘Verantwoord datadelen voor AI’ (‘Responsible data sharing for AI’) and the accompanying manual.
• Read the role that blockchain technologies can play in emerging data space architectures for trusted data sharing here.
• Data owners should also be able to control the secondary data generated for and by supporting data sharing processes, i.e. the metadata. Read more about how this can be designed in a data space environment here and here.
• The emerging data space architectures are leading to new interoperability challenges, particularly for legal agreements made for data sharing. Read more about this here.