The Per-Fide Corpus: : A New Resource for Corpus-Based Terminology, Contrastive Linguistics and Translation Studies
Bloomsbury Academic eBooks · 2014
The Per-Fide project is a joint collaboration between researchers at the Department of Informatics and the Institute of Arts and Humanities at the University of Minho, Portugal. The acronym Per-Fide stands for Portuguese (P) in parallel with 6 languages: English (E), Russian (R), French (F), Italian (I), German/Deutsch (D) and Spanish/ Espanol (E). First, we expound on the role of the Per-Fide project within the context of existing corpora that include the Portuguese language in its different variants – namely, European Portuguese, Brazilian Portuguese and Portuguese spoken in African countries (Angola, Mozambique, Guinea-Bissau, Cape Verde, Sao Tome and Principe). The idea of creating a multilingual parallel corpus project in which Portuguese assumes a pivotal role arose primarily due to the fact that the majority of online corpora that include Portuguese are either monolingual or bilingual. Furthermore, these corpora focus mainly on one specific text type. Consequently, the few multilingual parallel corpora that include Portuguese consist of a relatively small Portuguese subcorpus1 and provide limited search facilities mainly due to the fact that the Portuguese texts have not been morphologically tagged and/or syntactically annotated. Our second goal in this chapter is to provide an overview of the design criteria for the development of tools and resources in the various stages of the Per-Fide corpora construction process, focusing particularly on automation, validation, generalization and resource sharing. Here, a brief description of the workflow components involved in the preand post-alignment phases will be included. Finally, we draw attention to several practical applications of the current features of the Per-Fide corpus in translation practice and contrastive linguistic studies, focusing on the use and potential of probabilistic translation dictionaries and the role that parallel corpora can play in translating idioms.