Collocations in Parsing and Translation.

Proper identification of collocations (and more generally of multiword expressions (MWEs), is an important qualitative step for several NLP applications and particularly so for translation. Since many MWEs cannot be translated literally, failure to identify them yields at best inaccurate translation. This paper is mostly be concerned with collocations. We will show how they differ from other types of MWEs and how they can be successfully parsed and translated by means of a grammar-based parser and translator.

Automatic Annotation of French Medical Narratives with SNOMED CT Concepts.

Medical data is multimodal. In particular, it is composed of both structured data and narrative data (free text). Narrative data is a type of unstructured data that, although containing valuable semantic and conceptual information, is rarely reused. In order to assure interoperability of medical data, automatic annotation of free text with SNOMED CT concepts via Natural Language Processing (NLP) tools is proposed. This task is performed using a hybrid multilingual syntactic parser. A preliminary evaluation of the annotation shows encouraging results and confirms that semantic enrichment of patient-related narratives can be accomplished by hybrid NLP systems, heavily based on syntax and lexicosemantic resources.