Adriano Ferraresi

The WaCky Wide Web: A Collection of Very Large Linguistically Processed Web-Crawled Corpora

This article introduces ukWaC, deWaC and itWaC, three very large corpora of English, German, and Italian built by web crawling, and describes the methodology and tools used in their construction. The corpora contain more than a billion words each, and are thus among the largest resources for the respective languages. The paper also provides an evaluation of their suitability for linguistic research, focusing on ukWaC and itWaC. A comparison in terms of lexical coverage with existing resources for the languages of interest produces encouraging results. Qualitative evaluation of ukWaC versus the British National Corpus was also conducted, so as to highlight differences in corpus composition (text types and subject matters). The article concludes with practical information about format and availability of corpora and tools.

Old Needs, New Solutions: Comparable Corpora for Language Professionals

Use of corpora by language service providers and language professionals remains limited due to the existence of competing resources that are likely to be perceived as less demanding in terms of time and effort required to obtain and (learn to) use them (e.g. translation memory software, term bases and so forth). These resources however have limitations that could be compensated for through the integration of comparable corpora and corpus building tools in the translator’s toolkit. This chapter provides an overview of the ways in which different types of comparable corpora can be used in translation teaching and practice. First, two traditional corpus typologies are presented, namely small and specialized “handmade” corpora collected by end-users themselves for a specific task, and large and general “manufactured” corpora collected by expert teams and made available to end users. We suggest that striking a middleground between these two opposites is vital for professional uptake. To this end, we show how the BootCaT toolkit can be used to construct largish and relatively specialized comparable corpora for a specific translation task, and how, varying the search parameters in very simple ways, the size and usability of the corpora thus constructed can be further increased. The process is exemplified with reference to a simulated task (the translation of a patient information leaflet from English into Italian) and its efficacy is evaluated through an end-user questionnaire.

Building interpreting and intermodal corpora : a how-to for a formidable task

This contribution has a double aim. On the one hand, it highlights the various challenges and problems compilers of (simultaneous) interpreting and intermodal corpora are likely to face, and the solutions that were found and applied in three corpora of European Parliament plenary debates, i.e. EPIC, EPICG and EPTIC. On the other, it provides an accessible step-by-step guide for corpus developers who are working with European Parliament data, with the ultimate aim of bringing as far as possible into line the procedures used to transcribe the audio tracks, record metadata, annotate texts with part-of-speech and lemma information, perform text-to-text and text-to-audio/video alignment, and index the corpus for searching with appropriate corpus query tools. By adopting shared corpus building methods, it might be possible to leverage the substantial efforts already deployed by different research groups, and encourage others to take charge of new language pairs. This, we shall argue, might lead to a massively multilingual interpreting and intermodal corpus, through a distributed community effort.

Enhancing Machine Translation of Academic Course Catalogues with Terminological Resources

This paper describes an approach to translating course unit descriptions from Italian and German into English, using a phrase-based machine translation (MT) system. The genre is very prominent among those requiring translation by universities in European countries in which English is a non-native language. For each language combination, an in-domain bilingual corpus including course unit and degree program descriptions is used to train an MT engine, whose output is then compared to a baseline engine trained on the Europarl corpus. In a subsequent experiment, a bilingual terminology database is added to the training sets in both engines and its impact on the output quality is evaluated based on BLEU and postediting score. Results suggest that the use of domain-specific corpora boosts the engines quality for both language combinations, especially for German-English, whereas adding terminological resources does not seem to bring notable benefits.

Intermodal corpora and the translation classroom: what can translation trainers and trainees learn from interpreting?

Abstract This contribution focuses on didactic applications of intermodal corpora, i.e. corpora featuring interpreted and translated language. It relies on EPTIC, a multipletranslation and intermodal parallel corpus containing EU Parliament plenary speeches in Italian and English. The peculiar nature of EPTIC allows the investigation of a set of translational alternatives which are distinguished by modality- and task-based constraints (written vs. oral, translated vs. interpreted). To exemplify the potential of such corpus evidence, teaching activities focusing on collocations are proposed that encourage students to reflect on the decision-making processes involved in the slower-paced, reflective task of translation, vs. the faster, more automatic one of interpreting. A method is also described that can facilitate the selection of relevant didactic examples.