Current software used for translating text, while useful for the first draft – and a rough one at that – is nowhere near the level of properly “understanding” the language pairs being used, and producing a pleasantly flowing translation. The method of simply feeding a computer massive amounts of grammatical nuances and teaching it to use them correctly is an exceptionally difficult task.
Luckily, this crudely mechanistic technique is not the only game in town – the European Union had recently approved funding for two research projects that will use deep learning to achieve successful computerized translations between all of the 24 major languages used within its borders.
“This machine learning strategy has nothing to do with natural intelligence, but it does have similarities with the processes that occur in the human brain when we control the muscles in our bodies,” said project leader Josef van Genabith, Professor of Translation-Oriented Language Technologies at Saarland University and a Scientific Director at the German Research Centre for Artificial Intelligence (DFKI).
The first project, called QT21, is a collaborative effort undertaken by 14 leading research institutions for machine translation in Europe and Hong Kong that includes universities, research institutes and numerous private companies. The project will receive a total of 3.9 million euros and last for three years.
The second project, called the European Language Resources Coordination (or ELRC), has been contracted by the European Commission to collect suitable language data sets that will enable its automated translation platform (CEF AT) to be adapted and optimized for the daily requirements of public administrators in all EU Member States as well as Iceland and Norway. ELRC is one of the most comprehensive collections of language data worldwide.
“We are currently identifying all possible Language Resources covering related subject areas, such as texts and their translations from European government ministries in the areas of finance, economics, interior affairs and foreign affairs. These data sets help the European Commission to train the translation software and to adjust it to meet the requirements of public administrators and European citizens,” said van Genabith.
An explanation of how statistically driven machine translation works, featuring Genabith himself, can be viewed free of charge by clicking this link.
The results of research on machine translation will be presented at the annual meeting of the Association for Computational Linguistics (ACL) on August 7-12, 2016 at the Humboldt University in Berlin, Germany.