Break the Glass in OBDA

Proc. of the 23rd Int. Conf. on Modeling Decisions for Artificial Intelligence (MDAI). Lecture Notes in Artificial Intelligence. 2026.

Within the Ontology-Based Data Access (OBDA) framework, users can query a relational data source using an ontology to which the source is linked via declarative mappings. In a world where data sharing is widespread, ensuring privacy while managing data poses a significant challenge. Controlled Query Evaluation (CQE) is a privacy-preserving query answering framework in the presence of ontologies, where policies representing confidential information are used to devise suitable censors that enforce data protection. The integration of CQE within OBDA was recently proposed through the Policy-Protected OBDA (PPOBDA) framework, which is based on embedding policies into mappings. While PPOBDA provides a principled mechanism for enforcing privacy, it assumes fixed access controls that are not suitable for emergency scenarios requiring temporary elevation of data access privileges. In this work, we address this limitation by integrating the Break-the-Glass (BtG) mechanism into the PPOBDA framework. BtG is a well-established paradigm in access control that permits authorized users to override certain privacy policies under critical circumstances, while ensuring that such overrides are justified, authenticated, and fully auditable to prevent misuse. We formalize BtG enhanced PPOBDA by extending policy-protected mappings with BtG-specific mappings that are triggered under emergency conditions, and we implement and evaluate the resulting framework. Specifically, we adopt for the evaluation the well-known MIMIC-III hospital dataset, which has earlier been mapped, according to the OBDA framework, to the Fast Healthcare Interoperability Resources (FHIR) ontology. Our experiments show the practical implications of the framework, evaluating its applicability to real-world scenarios and the overhead introduced by BtG enabled policy management.

@inproceedings{MDAI-2026,
   title = "Break the Glass in OBDA",
   year = "2026",
   author = "Divya Baura and Diego Calvanese",
   booktitle = "Proc. of the 23rd Int. Conf. on Modeling Decisions for Artificial
Intelligence (MDAI)",
   publisher = "Springer",
   series = "Lecture Notes in Artificial Intelligence",
}