Knowledge that stays correct over time
Ontologies, terminologies and knowledge graphs change with every release. I build methods that detect those changes and repair the mappings and annotations that depend on them.
Lead Researcher in knowledge graphs and hybrid AI at the Luxembourg Institute of Science and Technology (LIST)
I build knowledge that machines can reason with, and keep it correct as the world it describes changes: medical terminologies, the language of historical testimonies, the skills of Europe's workforce.
Topics of my 102 papers, appearing in the year I first wrote about them
Hybrid AI that combines formal reasoning with statistical learning, applied where knowledge changes fast and errors are costly.
Ontologies, terminologies and knowledge graphs change with every release. I build methods that detect those changes and repair the mappings and annotations that depend on them.
Formal models of clinical guidelines, conflict detection for patients with several conditions, adaptive treatment plans, and the architecture of Luxembourg's eHealth platform.
How the meaning of words shifts across decades of documents, and how skills data can be shared and matched with training, with LLMs and symbolic AI working together.
I welcome collaborations, project proposals and PhD or postdoc applications.