Detecció de la relació malaltia-símptoma entre termes de l'àmbit mèdicuna aproximació basada en corpus

Abstract

Detection of the disease-symptom relation between terms in the medical sphere: a corpus-based approximation Experts in terminology state that specialised texts are, with respect to their content, structures formed by terms, and these terms are conceptually connected (Cabré, 1999). The main goal of this article is to prove the existence of linguistic patterns that show the disease–symptom relation. These patterns can be useful to detect the relation semiautomatically or automatically by means of a specific methodology. This methodology is based on the fact that, once we extract and analyse certain contexts, we can obtain linguistic marks that show this relation, allowing a generalisation of patterns. Our results show that most symptoms are not included in the definitions given in medical terminological resources, but they are indeed included in the medical corpus that has been analysed in this project. Moreover, most symptoms are described with specialised collocations.