Supplementary Material: From Privacy-Utility Trade-Offs to Policies

This dataset provides the Supplementary Material accompanying the paper "From Privacy-Utility Trade-Offs to Policies: Optimized Anonymization Recommendations for Data Trustees in Data Spaces" (DATA 2026, SciTePress). It benchmarks 11 spatial anonymization methods across three strategic categories (Traditional, Specialized, Hybrid) on a real-world dataset of 2,147 forest parcels in Baden-Württemberg, Germany (EPSG:25832). The material contains detailed privacy metrics, utility metrics, an adversarial threat model, parameter sensitivity analyses, and validated use-case-specific recommendations for data trustees operating within data spaces. Contents include: Overview of the 11 evaluated methods with key parameters Utility vs. privacy metric tables (displacement, area deviation, data reduction, vertex preservation) Adversarial threat model with three attack vectors (Homogeneity, Background Knowledge, Satellite Correlation) and a Combined Vulnerability Index Parameter sensitivity analyses (H3 resolution, Geohash precision / "Privacy Cliff", differential privacy budget ε) Qualitative failure-mode illustrations ("Sparse Forest", "Shaky Hand", "Patchwork") Validated use-case recommendations mapped to ODRL policies for Health, Manufacturing, and Mobility Data Spaces ISO 27001-based sensitivity classification matrix for cadastral forest attributes Related publication: Steinert, M., Nazarov, B., Reitz, T., & Tebernum, D. (2026). From Privacy-Utility Trade-Offs to Policies: Optimized Anonymization Recommendations for Data Trustees in Data Spaces. In Proceedings of the 15th International Conference on Data Science, Technology and Applications (DATA), Volume 2, pp. 1097–1108. SciTePress. DOI: 10.5220/0015215200004091