IXA Biomedical Translation System at WMT16 Biomedical Translation Task

Abstract

In this paper we present the system developed at the IXA NLP Group of the University of the Basque Country for the Biomedical Translation Task in the First Conference on Machine Translation (WMT16). For the adaptation of a statistical machine translation system to the biomedical domain, we developed three approaches based on a baseline system for English-Spanish and Spanish-English language pairs. The lack of terminology and the variation of the prominent sense of the words are the issues we have addressed on these approaches. The best of our systems reached the average of all the systems submitted in the challenge in most of the evaluation sets.

BSC Participation in the WMT Translation of Biomedical Abstracts

“…For that matter, we used the MetamorphoSys application provided by U.S. National Library of Medicine (NLM) to subset the language resources for our desired language pairs. Our approach is similar to what was proposed byPerez-de Viñaspre and Labaka (2016).Once the resource was available, we imported the MRCONSO RRF file to an SQL database to split the data in a parallel format in the two language pairs. Table2shows the number of parallel concepts for each pair.…”