Syntagmatic Behaviors of Verbs in Medical Texts: Expert Communication vs. Forums of Patients

In this paper, we propose an automatic contrastive analysis of the behavior of verbs, with regard to the semantic features of their arguments (subject, direct object, indirect object), within and across medical subcorpora. We compare four medical subcorpora with texts whose authors and intended readership have different levels of expertise. The semantic annotation of the subcorpora is based on semantic information provided by a medical terminology. Our results indicate that the proposed procedures and tools could be used for the automatic detection of different ways of expressing medical concepts and conceptual relations, according to the types of texts.