Show Less

Corpus Data across Languages and Disciplines


Edited By Piotr Pezik

Over the recent years corpus tools and methodologies have gained widespread recognition in various areas of theoretical and applied linguistics. Data lodged in corpora is explored and exploited across languages and disciplines as distinct as historical linguistics, language didactics, discourse analysis, machine translation and search engine development to name but a few. This volume contains a selection of papers presented at the 8 th edition of the Practical Applications in Language and Computers conference and it is aimed at helping a wide community of researchers, language professionals and practitioners keep up to date with new corpus theories and methodologies as well as language-related applications of computational tools and resources.


Show Summary Details
Restricted access

Translation Quality Checking in Language Tool: Marcin Miłkowski


Translation Quality Checking in LanguageTool Marcin Miłkowski Abstract In large computer-aided translation (CAT) projects, especially in software localization, one of the main problems is to maintain the consistent style of the translated text. To tackle this problem, translators have to follow different guidelines defined in style guides for different translation jobs. Yet, in the case of conflicting guidelines (for example, terminological) for various projects it is very easy to make mistakes, and quite hard to find them because they are neither obvious nor glaring errors. Automated translation quality assessment (QA), on the other hand, are usually quite costly compared to other CAT tools and/or do not have any comprehensive natural-language processing features. Usually, their use is not really beneficial for languages other than English. Because of that, the proofreading process is costly and time-consuming, or the translation quality is negatively impacted. In this paper, I will present the translation QA features available in LanguageTool, an open- source proofreading tool (Miłkowski 2010) LanguageTool currently (as of version 1 4 released on June, 26, 2011) supports 22 languages and is able to use the standard target language rules to check for the mistakes in the translated text, including false friends in translation, as well as specially designed translation QA rules. These rules may be specially crafted to conform to informal style guides and include the most frequent mistakes found by human proofreaders (using the method specified in Miłkowski, forthcoming) I will show some examples of such bilingual rules. It...

You are not authenticated to view the full text of this chapter or article.

This site requires a subscription or purchase to access the full text of books or journals.

Do you have any questions? Contact us.

Or login to access all content.