Automated extraction of food-drug interactions from scientific articles
Résumé
In this paper, we are interested in the extraction of food-drug interactions (FDI), a task which is similar to the extraction of relation between terms in specialized texts. We present a supervised classification method and the results of a first set of experiments. Despite the imbalance of classes, the results are encouraging. We have identified the most relevant classifiers according to the steps of our method. We have also observed the important impact of the semantic tags of terms used as features.
| Origine | Fichiers produits par l'(les) auteur(s) |
|---|---|
| Licence |