Experiments in synonymy: weakly supervised term matching to concepts

Généreux, Michel; Mendes, Amália; Hamon, Thierry

http://hdl.handle.net/10451/37410

Utilize este identificador para referenciar este registo.

Nome:	Descrição:	Tamanho:	Formato:
tia2013-u8-final.pdf		141.32 KB	Adobe PDF	Ver/Abrir

Contacte-nos

Autores

Généreux, Michel

Mendes, Amália

Hamon, Thierry

Resumo(s)

We describe experiments using distributional semantics to extract and map simple terms from a corpus of genomics in Portuguese. A list of salient terms is first extracted from the corpus and manually verified by a geneticist. We then suggest a list of possible matches for the salient terms among the Unified Medical Language System (UMLS) semantic types and concepts using a Vector Space Model (VSM). Experiments show that salient terms can be extracted efficiently, while mapping them to specific concepts from a large set semantic types proved much more difficult.

URI

http://hdl.handle.net/10451/37410

Citação

Généreux, M., Mendes, A. & Hamon, T. (2013): “Experiments in synonymy: weakly supervised term matching to concepts”, Proceedings of the 10th International Conference on Terminology and Artificial Intelligence, Paris, 28 - 30 October 2013, pp. 181-184

Editora

Université Paris

Coleções

FL - CLUL - Livros de Actas

Ver registo completo