The Scielo Corpus: a Parallel Corpus of Scientific Publications for Biomedicine

The Scielo Corpus: a Parallel Corpus of Scientific Publications for Biomedicine Mariana Neves∗, Antonio Jimeno Yepes†, Aur´elie N´ev´eol‡ ∗Hasso-Plattner Institute August-Bebel-Str. 88 14482 Potsdam - Germany mariana.neves@hpi.de †IBM Research Australia, Carlton 3053 VIC - Australia, University of Melbourne, Parkville 3010 VIC - Australia antonio.jimeno@au1.ibm.com ‡LIMSI, CNRS, UPR 3251 Universit´e Paris-Saclay F-91405 Orsay - France aurelie.neveol@limsi.fr Abstract The biomedical scientific literature is a rich source of information not only in the English language, for which it is more abundant, but also in other languages, such as Portuguese, Spanish and French. We present the first freely available parallel corpus of scientific publications for the biomedical domain. Documents from the ”Biological Sciences” and ”Health Sciences” categories were retrieved from the Scielo database and parallel titles and abstracts are available for the following language pairs: Portuguese/English (about 86,000 documents in total), Spanish/English (about 95,000 documents) and French/English (about 2,000 documents). Additionally, monolingual data was also collected for all four languages. Sentences in the parallel corpus were automatically aligned and a manual analysis of 200 documents by native experts found that a minimum of 79% of sentences were correctly aligned in all language pairs. We demonstrate the utility of the corpus by running baseline machine translation experiments. We show that for all language pairs, a statistical machine translation system trained on the parallel corpora achieves performance that rivals or exceeds the state of the art in the biomedical domain. Furthermore, the corpora are currently being used in the biomedical task in the First Conference on Machine Translation (WMT’16). Keywords: parallel corpus, machine translation, biomedicine 1. Introduction Researchers make available new findings and knowledge in biology and medicine through the publication of research articles. However, the volume of the biomedical literature grows at a large rate, which poses a challenge to keep up with new discoveries. Even though a large volume of pa- pers are published in English, there are also many publica- tions written in languages other than English. Access to the biomedical literature is available on-line via systems such as PubMed®1, that allow researchers to browse and search for publications of their interest. Even though English is the de facto official language in the scien- tific community, many researchers have limited proficiency in English and feel more comfortable with reading scien- tific prose written in their native language. The same goes for patients who might find it difficult to understand medi- cal records when they are written in a language other than their native language. Furthermore, articles published in lo- cal research journals are accessible only for researchers flu- ent in the original language of the article. This is typically the case for articles available in databases such as Scielo2, which has a focus on Latin American publications. Machine translation (MT) can provide a solution to in- crease the access to the biomedical literature (Pecina et al., 2014) and to health information in general (Kirchhoff et al., 2011). For instance, it was successfully used as the ba- 1http://www.ncbi.nlm.nih.gov/pubmed 2http://www.scielo.org/ sis for query translation in …

Content truncated to 500 words