Leveraging NLP Technologies and Linked Open Data to Create Better CAT Tools

This paper presents a prototype of a Computer Aided Translation (CAT) interface integrated with an entity extraction system to create a dynamic linked terminology component. The entity extraction system tags terms in the source sentence, mapping them to translation candidates in the target language. A usage scenario for linked data within a CAT tool is evaluated by prototyping all components necessary to construct a real-time dynamic terminology. By making use of Natural Language Processing (NLP) technologies including entity linking (Mihalcea et al. 2007), and statistical models for extracting and disambiguating entities (Daiber et al. 2013), the tool can provide translators with rich feedback about potential target-language translations of entities in the source text.