Selection of relevant articles for curation for the Comparative Toxicogenomic Database

TitleSelection of relevant articles for curation for the Comparative Toxicogenomic Database
Publication TypeConference Proceedings
Year of Conference2012
AuthorsVishnyakova, D, Pasche, E, Ruch, P
Conference Name2012 BioCreative Workshop
Pagination31-38
Conference LocationWashington, DC USA
Abstract

We report on t he original integration of an automatic text categorization pipeline, so-called ToxiCat (Toxicogenomic Categorizer) to perform biomedical documents classification and prioritization in order to speed up curation of the Comparative Toxicogenomics Database (CTD).
The task can be basically described as a binary classification task, where relevance scores are used to ranks a selected set of articles. We design a SVM classifier, which combines four main
components: an information retrieval engine for MEDLINE (EAGLi), a biomedical named-entity recognizer based on terminological resources, a gene normalization (GN) service (NormaGene) developed for a previous BioCreative campaign and finally, an ad-hoc keyword recognizer for diseases and chemicals. The main components of the pipeline are publically-available both as web application and web services. The integration performed for the BioCreative competition is
available via a web user-friendly interface: http://pingu.unige.ch:8080/Toxicat.

URLhttp://www.biocreative.org/media/store/files/2012/Proceedings_BC2012_.pdf