---

I am associate professor (Maître de Conférences) in Computer Science department at the Paris-Nord university. I do my research work at the Medical Informatics and Bioinformatics Lab (LIM&BIO - EA3969 - Université Paris Nord) and my teaching at the Galilée institute.

---

News:

• (Mai 2013) Acceptation of a article in Patient Education and Counseling
• (May 2013) Acceptation of a paper ASMDA 2013
• (April 2013) Acceptation of a paper at TALN 2013
• (April 2013) Acceptation of a paper (poster presentation) at MedInfo 2013
• (April 2013) Acceptation of a paper at AIME 2013
• (April 2013) Acception of an article in Biomedical Informatics Insights
• (April 2013) Publication of an article in JAMIA
• (March 2013) BioYaTeA a version of YaTeA dedicated to the term extraction in biological text is available. BioYaTeA (mainly developed by the Bibliome team, MIG-INRA) is a supporting resource of shared Task BioNLP2013.
• (February 2013) A new version of the term extractor YaTeA (Version 0.622 is online)

---

My research activity addresses the Natural Language Processing (NLP) and the corpus linguistics, mainly in specialised areas (medicine, biology, agronomy, telecomunication, etc). I am interested by the terminology extraction and structuring.

My research work aims at developing approaches based on semantic information, and building specialised resources (thesaurii, specialised lexicons). I argue that such resources can not be viewed as term lists but must be structured and also provide relations between terms i.e. morphological and semantic ones, and term classes.

Additionally to the work on terminology structuring, I am also interested in combining linguistic tools in a processing text line. Since a decade, several NLP tools are available for linguisitic analysis of textual documents: word and sentence segmenatiser, named entity recogniser, lemmatiser, Part-Of-Speech tagger, term extractor, semantic relation identifier. Even if the quality of the results are variables, such information are complementary. Managing such information is not obvious. Currently, I focus on the implementation aspects while developing a distributed architecture for linguistic annotation of specialised documents.