New cross-lingual workflow transfer for large-scale content
Language translation is a complex endeavour. It is hard to find a one-to-one equivalent when transferring content from one language to another as each language has its own system for conveying concepts. The EU-funded WIKOLLECT project will explore this issue by drawing on a synergy between natural language processing, language learning, and crowdsourcing. It will develop a special workflow for large-scale transference of high-quality content across languages. This includes four cyclic steps to automatically identify content in the source language that is missing in the target language and generate potential translations. Applied in Italian and German to Wiktionary, the free-content multilingual online dictionary, this project workflow will promote the fair re-use of content across languages and facilitate knowledge transfer.
Fields of science
Call for proposal
See other projects for this call