Category: SIESTA Concepts
SIESTA Concepts #5: Named Entity Recognition
Sharing cyber incident reports is vital for threat detection, but they’re full of personal data. Anonymisation pipelines rely on Named Entity Recognition models to catch sensitive mentions first — and these are usually trained in English. Researchers within EOSC SIESTA tested whether that holds up in Spanish, and found that multilingual models outperform English cybersecurity…
SIESTA Concepts #4 | FAIR Data
EOSC-SIESTA bridges the gap between open science and data protection by applying FAIR principles to sensitive data. Through trusted cloud environments, the project demonstrates that data can be secure, confidential, and controlled while remaining findable, accessible, interoperable, and reusable for scientific research.
Siesta Concepts: Open Science
Open Science promotes a transparent and collaborative research ecosystem, driven by Europe’s EOSC initiative. However, sharing sensitive data requires balancing openness with privacy. EOSC-SIESTA addresses this challenge by developing trusted, cloud-based environments and privacy-preserving tools that make sensitive data securely shareable and usable under FAIR principles.
Siesta Concepts #2: Anonymization
Data anonymization transforms sensitive datasets to prevent individual identification while keeping data useful. Within EOSC-SIESTA, two open-source tools help researchers sanitize quasi-identifiers and assess privacy risks by applying advanced anonymity models (like k-anonymity and l-diversity) to protect sensitive tabular data.
SIESTA CONCEPTS#1: Federated Learning
Federated Learning (FL) allows multiple participants to collaboratively train AI models by keeping data local and sharing only encrypted model updates. Through the SIESTA dashboard, users can easily deploy secure FL clients, which connect to a central coordinating server hosted on AI4EOSC via the Flower framework and secure gRPC protocols, enabling privacy-preserving collaboration.