Since August 2020, I am an AI Specialist (expert in NLP) at the Alexandra Institute. I am working on building NLP tools for the Danish language in collaboration with Universities and companies through the "Dansk for alle" project.
For more than 2 years, I was a Data Scientist at Siteimprove where I pursued research in Natural Language Processing and developed tools that help our customers maintain and improve their websites. In particular, I worked on developing a grammatical error correction tool and supervised a Ph.D. student in the same field.
Before that, I was a postdoctoral researcher at the Department of Computer Science of the University of Copenhagen in the CoAStaL group and worked on the LOWLANDS research project (ERC). My research focused on cross-lingual transfer of information between languages, in particular for low-resource languages and domains, through tasks such as PoS-tagging and discourse segmentation.
Even before, I was a postdoctoral researcher at the LIMSI-CNRS where I joined the TLP (Spoken Language Processing) group and I took part in the project PAPYRUS. My research focused on dependency parsing and in particular on learning dependency parsers from partially annotated data. Through this research, I have been interested in cross-lingual transfer of syntactic information, active learning, constrained learning/parsing and domain adaptation. I was also involved in the Automatic translation and machine learning group in which I worked on word pre-ordering for machine translation and in the use of dependency syntax for translating morphologically rich languages.
I am a Ph.D. in computational linguistics of the University of Nantes in France. I defended my thesis in December 2014. The subject of my thesis focused on non-projective dependency parsing, including the use of statistical methods through the study of grammar-based and transition-based parsers.