Nlp Meets Library Science: Providing a Set of Enhanced Language Reference Tools for Online Translators

Introduction. We are developing an online translation aid tool that provides enhanced language reference tools, specifically designed for translators working online. This paper introduces the concepts and framework for enhancing and organising the reference tools to maximally help translators. Method. This paper is an essentially theoretical work, resorting to deduction from basic premises on language and translation. In order to design a maximally useful system, however, we also analysed translators’ use of reference tools and examined the social arrangement of language reference tools. Results. The main tenet of this paper is that existing natural language processing (NLP) technologies can be useful but for them to be actually used by language practitioners (translators in this context), it is necessary to take into account the social arrangement of the reference tools – the sort of knowledge that library science has accumulated. We have developed the Shiitake/QRedit translation aid system, in which a range of enhanced language reference sources are provided in a stratified and organised way that reflects the essential elements in the process of translation and the social arrangement of reference tools. Conclusion. We have provided a sound basis for the design of the translation aid system and shown a way for NLP technologies to be fully utilised in real-world situations. The system we have developed is in experimental use (due to copyright restrictions on the core dictionary) by translators and has been enthusiastically received.