Research

Knowledge Discovery - Big Data Mining

To keep up with and go beyond state-of-the-art, knowledge workers must extract and interpret valuable and relevant knowledge from the huge amount of documents available in multiple repositories. In life science, they also need to mine the big data sets they collect from the lab experiments.
In this context, I am interested in semantic infrastructures supporting biocuration for genomics and biomedical research, as well as machine-learning-based classification systems capable of detecting relevant documents in highly imbalanced corpora. In parallel with literature mining, I design data mining and analysis tasks based on relational and statistical approaches.

Graph Theory

I am also interested in graph theory and its application to linked open data and information retrieval. I specifically studied the Vertex Separator Problem (Read more on this problem here and there).
Vertex Separator Problem

Spoken Dialog Systems

From 2006 to 2010, I was a member of the Human/Machine Dialog group of the team at LIA, University of Avignon.
My research interests were focused on probabilistic models for spoken language interpretation in human/machine dialogs in the context of the LUNA FP6 European project.