The anonymisation process involves turning personal or sensitive information (e.g., health data) to anonymous information, by removing personal identification information from a dataset.The anonymisation tool DANS 2.0 (Data Anonymisation Service 2.0) is intended to protect and preserve the privacy of personal and sensitive information by removing or modifying identifiable information avoiding the re-identification of the data subject, ensuring that the anonymisation process will maintain the utility of the protected data for further analysis. DANS 2.0 is based on the legacy DANS asset, and those provided by the Amnesia open-source library, providing data anonymisation techniques such as generalisation, suppression and micro aggregation, and supporting different privacy models .e k-anonymity, l-diversity, t-closeness, km-anonymity or, differential privacy. These techniques and models will be applied to different types of data such as Electronic Health Records (EHRs).

For facilitating the implementation and deployment processes, the design and development of this asset follows a modular design based on a microservice approach. This modular design will support the scaling-up needs when big data analysis is required. In the context of the SECURED project, DANS2.0 will be offered as a tool, to be deployed on the data provider infrastructure. A docker image contains the different modules. An OpenAPI is provided for facilitating their integration on wider frameworks.

The anonymised file will be checked by the user for privacy/utility trade-off. If the user considers the anonymisation fulfills their initial expectations, they can use the de-anonymisation tool for a deeper analysis of the tentative attacks the anonymised dataset can suffer.Otherwise, the user can use different privacy models or parameters for additional anonymisation processes.