DATAPACT: COMPLIANCE BY DESIGN OF DATA/AI OPERATIONS AND PIPELINES

Digital, Industry & SpaceHORIZON-IAID: 101189771
EC Contribution
€100,000
Consortium Size
19 orgs
Start Year
2025
Summary

DataPACT develops novel tools and methodologies that enable efficient, compliant, ethical, and sustainable data/AI operations and pipelines. DataPACT delivers a transformative approach where compliance, ethics, and environmental sustainability are not afterthoughts but foundational elements of data/AI operations and pipelines. DataPACT contributes to the design, implementation, and management of data/AI operations and pipelines by embedding compliance, privacy, and environmental sustainability at their core design. It delivers compliance by design for data/AI operations and pipelines by developing innovative technical tools (DataPACT Compliance Toolbox) and tool-supported methodologies (DataPACT Compliance Framework) for compliance assessment and realization of data/AI pipelines designed, deployed and executed through a set of pipeline management tools and techniques (DataPACT Compliance-aware Data/AI Pipeline Toolbox).DataPACT validates its core results through a strong selection of seven complementary use cases offered by SMEs, large companies, and public sector organizations in relevant areas, including media and entertainment, healthcare, smart cities, law enforcement and security, customer relationship management, manufacturing, and public data. DataPACT helps them ensure straightforward and cost-effective compliance with existing and emerging regulations and guidelines, shorter time-to-market for compliant data solutions, fair and unbiased data-driven systems, respect for privacy and other fundamental rights, and lower and transparent environmental impact for intensive data/AI pipeline operations.DataPACT gathers a balanced consortium of 18 partners from 16 countries, consisting of two large companies, two public sector organizations, four SMEs, two research centers, and seven universities, covering relevant aspects related to legal, ethical, social, environmental, and technical compliance of data/AI operations and pipelines.

Consortium (19)

Project Results (24)

Source: CORDIS, the EU research results database.

Publications (18)
ADApt: Edge Device Anomaly Detection and Microservice Replica Prediction
2025 IEEE 9th International Conference on Fog and Edge Computing (ICFEC)· 2025DOI
Mehran, Narges; Nikolov, Nikolay; Prodan, Radu; Roman, Dumitru; Kimovski, Dragi; Pallas, Frank; Dorfinger, Peter
Bag Containment of Join-On-Free Queries.
· 2025DOI
Konstantinidis, George; Mogavero, Fabio
DAVE: A Framework for Assisted Analysis of Document Collections in Knowledge-Intensive Domains
Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence· 2025DOI
Ruben Agazzi, Renzo Alva Principe, Riccardo Pozzi, Marco Ripamonti, Matteo Palmonari
How CO2STLY Is CHI? The Carbon Footprint of Generative AI in HCI Research and What We Should Do About It
Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems· 2025DOI
Nanna Inie; Jeanette Falk; Raghavendra Selvan
IEEE Access
IEEE Access· 2025DOI
Inna Novalija; Dumitru Roman; Federico Belotti; Vladimir Alexiev; Luis Rei; Roberto Avogadro; Babak Khalilvandian; Boyan Bechev; Catalina Alexandra Chinie; Iulia Ciurea; Janez Brank; Cosmin Udroiu; Ahmet Soylu; Matteo Palmonari
MammoTab 25: A Large-Scale Dataset for Semantic Table Interpretation - Training, Testing, and Detecting Weaknesses
Lecture Notes in Computer Science, The Semantic Web – ISWC 2025· 2025DOI
Marco Cremaschi, Federico Belotti, Jennifer D’Souza, Matteo Palmonari
Prompt2DAG: A Modular Prompting Approach for Democratizing Data Pipeline Generation
2025 IEEE International Conference on Software Services Engineering (SSE)· 2025DOI
Abubakari Alidu, Michele Ciavotta, Flavio De Paoli
ReFactX: Scalable Reasoning with Reliable Facts via Constrained Generation
Lecture Notes in Computer Science, The Semantic Web – ISWC 2025· 2025DOI
Riccardo Pozzi, Matteo Palmonari, Andrea Coletta, Luigi Bellomarini, Jens Lehmann, Sahar Vahdati
Towards a Toolkit for Semantic Interoperability in Data Spaces
2025 IEEE 8th International Conference on Industrial Cyber-Physical Systems (ICPS)· 2025DOI
An Ngoc Lam, Roberto Avogadro, Francisco Martin-Recuerda, Brian Elvesæter, Xiang Ma, Erik Johan Nystad, Dumitru Roman, Arne J. Berre
Automated Compliance for Data and AI Pipelines: the DataPACT Project
• Maidens, C., Pareti, P., Roman, D., & Konstantinidis, G.
Climate And Resource Awareness is Imperative to Achieving Sustainable AI (and Preventing a Global AI Arms Race)
Pedram Bakhtiarifard, Pınar Tözün, Christian Igel, Raghavendra Selvan
Efficient Uncertainty Estimation for LLM-based Entity Linking in Tabular Data
Carlo Bono, Federico Belotti, Matteo Palmonari
Evaluation and Comparison Semantics for ODRL
1. Salas, J. O., Pareti, P., Yumusak, S., Gheisari, S., Ibáñez, L.-D., & Konstantinidis, G.
Group-SAE: Efficient Training of Sparse Autoencoders for Large Language Models via Layer Groups
Davide Ghilardi, Federico Belotti, Marco Molinari, Tao Ma, Matteo Palmonari
Knowledge graph-driven policy enforcement in urban dataspace
12. Petrova-Antonova, D., Pareti, P., Tomov, P., Yumusak, S., Maidens, C., Jiang, S., Konstantinidis, G., & Roman, D.
MINiDM: Multi-issue negotiation in decentralised data marketplaces
14. Gheisari, S., Salas, J. O., Yumusak, S., & Konstantinidis, G.
Oscillations make neural networks robust to quantization
Jonathan Wenshøj, Bob Pepin, Raghavendra Selvan
Performance per Resource Unit as a metric to promote Small-Scale Deep Learning in Medical Image Analysis
Raghavendra Selvan, Bob Pepin, Christian Igel, Gabrielle Samuel, Erik B Dam
Deliverables (6)