Proposal for an automatic extraction for medical term candidates processing linguistic information. Description and evaluation of results

• Autor(a) principal: Koza Orellana, Walter
• Data de Publicação: 2015
• Tipo de documento: Artigo
• Idioma: por; eng
• Título da fonte: Alfa (São José do Rio Preto. Online)
• Texto Completo: https://periodicos.fclar.unesp.br/alfa/article/view/6440
• topic: Medical terminology; Automatic extraction; Linguistic information; Terms candidate; Terminología médica; Extracción automática; Información lingüística; Candidatos a término

The description of a method for automatic extraction of term candidates from the medical field by applying linguistic information is presented. Lexicography, morphological and syntactic rules were used. First, the detection was performed by applying a standard dictionary that assigned the tag ‘MED’ (‘MEDICAL’) to the words that could be considered terms. Morphological and syntactic rules were used to try to deduce the part of speech of the words that were not considered in the dictionary (WNCD). Afterwards, nominal phrases that included WNCD and MED were gathered to extract them as term candidates of the field. Smorph, Post Smorph Module (MPS) – both working in groups – and Xfst were the software used. Smorph performs the morphological analysis of character strings and MPS works on local grammar. Xfst is a finite state tool that works on character strings assigning previously stated categories to allow the automatic analysis of expressions. This method was tested on a section of the corpus of clinical cases collected by Burdiles (CCCM - 2009) containing 217,258 words. The results showed 92.58% of precision, 95.02% of recall and 93.78% of F-measure.