Working with Corpora in Translation Technology Teaching: Enhancing Aspects of Course Design

This chapter offers a discussion of two linked areas that are relevant to the translation technology curriculum. Firstly, in the absence of a suitable cost-effective industry-standard terminology extraction utility—as well as the relative under-representation of this feature in most CAT tools—the Sketch Engine lexical analysis suite of tools has provided our students with a reasonably successful application to use for this purpose. Sketch Engine permits users to create monolingual lists of single- and multi-word candidate terminological items on the basis of ad hoc corpora created from texts that need to be translated, these lists then providing students with a solid starting point from which to create detailed bilingual terminology lists, while a bilingual term extraction function is also provided. Secondly, aware that many CAT tool functions work best when large translation memories are available, but also mindful of the fact that students generally have to start with TMs that are completely empty, we use sources such as the OPUS website as a resource for downloading large corpora to facilitate the creation of very sizable translation memories with great rapidity. In this way, students are able to obtain a much more positive impression of a CAT tool’s concordance tool, for example, and also potentially gain confidence in the tool’s capacity to provide useful fuzzy matches in general.