I am currently working as a Postdoctoral Visiting Fellow in Zhiyong Lu‘s Biomedical Text Mining group at the National Center for Biotechnology Information of the National Institutes of Health, Bethesda, USA. I also collaborate during my free time with Moreno Mitrović on topics related to computational linguistics. In the past – still very recently to me –, I was a PhD student in the KID’s team at LGI2P lab of École des Mines d’Alès at Nîmes, France. I happened to work with the MAB team of the LIRMM at Montpellier, on phylogenetics and evolution. On these latter topics, I also worked with the Ensembl team of the European Bioinformatics Institute on genomic alignments.

Current project

As a researcher at the NCBI, I am privileged to be able to work on the PubMed information retrieval system. While it has proven to be very efficient, it does not make use of the state-of-the-art methods in Information Retrieval. My work focuses on improving the current system by providing more accurate query answers through a better understanding of queries, documents and relevance. To this aim, I make use of classical and/or novel techniques from the fields listed below. Hopefully, this work will improve the experience of the millions of users PubMed welcomes every day.

Research interests

  • machine learning
  • information retrieval
  • deep learning
  • text mining
  • health informatics
  • natural language processing
  • algorithmics & optimization
  • recommender systems
  • knowledge bases
  • semantic similarity
  • conceptual indexing