Pierre Zweigenbaum
Tutorial:
Detecting Medical Concepts in Clinical Texts (named entity extraction and use of specialized vocabularies, terminologies, ontologies)
- Terminological and ontological resources for biomedical language processing (structure of biomedical terminologies, introduction to a subset of terminological resources including MeSH, SNOMED and UMLS, dictionary-based methods for biomedical entity detection) - 45' (Pierre 1)
- Biomedical information extraction : entity detection and relation extraction (supervised entity detection methods, supervised relation detection, examples from biomedical shared tasks) - 1h30 (Pierre 2, Pierre 3)
- Hybrid biomedical IE methods (combining dictionaries and machine learning) - 45' (Pierre 4)
Pierre and Eric will combine their tutorials, as follows:
Pierre 1 (45’), Éric 1 (45’)
Éric 2 (45’), Pierre 2 (45’)
Pierre 3 (45’), Éric 3 (45’)
Pierre 4 (45’), Éric 4 (45’)
Hands-on exercises:
analysis and discussion on (some) information extraction tools for biomedical texts
(together with Eric)
- Session 1 (90’): Tools for information extraction in the biomedical domain
- Session 2 (90’): Development of a classifier for concept extraction in short clinical texts
Short bio:
Pierre Zweigenbaum is a researcher at LIMSI, CNRS, Université Paris-Saclay, Orsay (France)
