Skip to Main content Skip to Navigation
Conference papers

Fouille de motifs et CRF pour la reconnaissance de symptômes dans les textes biomédicaux

Abstract : Pattern mining and CRF for symptoms recognition in biomedical texts. In this paper, we tackle the issue of symptoms recognition in biomedical texts. There is not much attention to this problem in the literature and it does not exist to our knowledge an annotated dataset to train a model. We propose two weakly-supervised approaches to extract these entities. The first is based on pattern mining and introduces a new constraint based on semantic similarity. The second represents the task as sequence labeling using CRF (Conditional Random Fields). We describe our experiments which show that the two approaches are complementary in terms of quantification (recall and precision). We further show that their combination significantly improves the results.
Document type :
Conference papers
Complete list of metadata

Cited literature [16 references]  Display  Hide  Download
Contributor : Laboratoire MoDyCo Connect in order to contact the contributor
Submitted on : Monday, March 12, 2018 - 2:11:15 PM
Last modification on : Saturday, June 25, 2022 - 9:51:50 AM
Long-term archiving on: : Wednesday, June 13, 2018 - 12:13:01 PM


Publisher files allowed on an open archive


  • HAL Id : halshs-01727081, version 1


Pierre Holat, Nadi Tomeh, Thierry Charnois, Delphine Battistelli, Marie-Christine Jaulent, et al.. Fouille de motifs et CRF pour la reconnaissance de symptômes dans les textes biomédicaux. 23e conférence sur le Traitement Automatique des Langues Naturelles (TALN’16), Jul 2016, Paris, France. pp.194-206. ⟨halshs-01727081⟩



Record views


Files downloads