MuEVo, a breast cancer Consumer Health Vocabulary built out of web forums

Solène Eholié 1 Mike Donald Tapi Nzali 2, 1 Sandra Bringay 3, 1 Clement Jonquet 4
1 ADVANSE - ADVanced Analytics for data SciencE
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier
4 SMILE - Système Multi-agent, Interaction, Langage, Evolution
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier
Abstract : Semantically analyze patient-generated text from a biomedical perspective is challenging because of the vocabulary gap between patients and health professionals. The medical expertise and vocabulary is well formalized in standards terminologies and ontologies, which enable semantic analysis of expert-generated text; however resources which formalize the vocabulary of health consumers (patients and their family, laypersons in general) remain scarce. The situation is even worse if one is interested in another language than English. In previous studies, we attempted to produce a French preliminary Consumer Health Vocabulary (CHV) by mining the language used within online public forums & Facebook groups about breast cancer. In this work, we show our effort to concretely align the vocabulary produced to standard terminologies and to represent its content (terms & mappings) using semantic web languages such as RDF, SKOS and PROV. We used a sample of 173 relations built around 64 expert concepts which have been automatically (89%) or manually (11%) aligned to standard biomedical terminologies, in our case: MeSH, MedDRA and SNOMEDint. The resulting vocabulary, called MuEVo (Multi-Expertise Vocabulary) and the mappings are publicly available in the SIFR BioPortal French biomedical ontology repository.
Type de document :
Communication dans un congrès
International Semantic Web Applications and Tools for Life Sciences, SWAT4LS, Dec 2016, Amsterdam, Netherlands
Liste complète des métadonnées

Littérature citée [24 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-01583558
Contributeur : Mike Donald Tapi Nzali <>
Soumis le : jeudi 7 septembre 2017 - 15:25:27
Dernière modification le : jeudi 26 octobre 2017 - 13:44:16

Fichier

Article-SWAT4LS-2016_MUEVO.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01583558, version 1

Citation

Solène Eholié, Mike Donald Tapi Nzali, Sandra Bringay, Clement Jonquet. MuEVo, a breast cancer Consumer Health Vocabulary built out of web forums. International Semantic Web Applications and Tools for Life Sciences, SWAT4LS, Dec 2016, Amsterdam, Netherlands. 〈hal-01583558〉

Partager

Métriques

Consultations de la notice

36

Téléchargements de fichiers

9