Soutenance de thèse Ashley Caselli
 |
M. Ashley Caselli soutiendra en anglais, en vue de l'obtention du grade de docteur ès sciences de la société, mention systèmes d'information, sa thèse intitulée: A Domain-Agnostic Semantic Framework for Automating Regulatory Compliance Date: Lundi 16 juin 2025 à 9h00 |
Jury :
- Prof Giovanna Di Marzo Serugendo — supervisor, UNIGE
- Prof Gilles Falquet — co-supervisor, UNIGE
- Prof Jean-Henry Morin — president, UNIGE
- Lamia Friha — Responsable Cellule R&D/Lecturer/Responsable l'Accélérateur de sciences et services numériques - DiSTIC (UNIGE)
- Livio Robaldo — Senior Lecturer in Computational Law @ Swansea University
- Holger Knublauch — Lead Software Developer @ TopQuadrant
Abstract:
Organizations face significant challenges in ensuring compliance with complex and ever-growing legal and regulatory requirements, typically written in natural language. This makes automation difficult and resource-intensive. Existing compliance solutions are often domain-specific and lack scalability and reusability. This research proposes a domain-agnostic framework leveraging Semantic Web technologies to automate regulatory compliance processes. These technologies enable semantic data enrichment, interoperability, and the creation of FAIR-compliant knowledge bases.
The framework introduces a unified model for regulatory knowledge formalization, implemented as an ontology-based language. It separates domain-specific business logic from regulatory representations, promoting maintainability and reusability. Additionally, it outlines an abstract workflow for both compliance checking and compliance-by-design approaches, adaptable to various regulatory contexts.
A proof-of-concept (PoC) demonstrates the framework’s flexibility and effectiveness across multiple domains, including (i) architecture, engineering, and construction (AEC), (ii) finance, and (iii) academia, using knowledge graphs and SHACL. Results show the framework’s potential to streamline compliance efforts and support organizations and practitioners in managing regulatory knowledge more efficiently. Future work will refine the framework and expand its application scope, with the goal of further promoting automation, scalability, and traceability of regulatory compliance processes across industries.
The framework introduces a unified model for regulatory knowledge formalization, implemented as an ontology-based language. It separates domain-specific business logic from regulatory representations, promoting maintainability and reusability. Additionally, it outlines an abstract workflow for both compliance checking and compliance-by-design approaches, adaptable to various regulatory contexts.
A proof-of-concept (PoC) demonstrates the framework’s flexibility and effectiveness across multiple domains, including (i) architecture, engineering, and construction (AEC), (ii) finance, and (iii) academia, using knowledge graphs and SHACL. Results show the framework’s potential to streamline compliance efforts and support organizations and practitioners in managing regulatory knowledge more efficiently. Future work will refine the framework and expand its application scope, with the goal of further promoting automation, scalability, and traceability of regulatory compliance processes across industries.