'Using linguistic information for hybrid machine translation', Barcelona, Spain
18 November 2011 - 18 November 2011
Spain
An event entitled 'Using linguistic information for hybrid machine translation' will take place on 18 November 2011 in Barcelona, Spain.
Data-driven machine translation is increasingly used prevalent within the machine translation research community. From a quality point of view, results are becoming increasingly useful, especially when the target language is English.
Data-driven machine translation systems base their knowledge on bilingual aligned corpora, which are large and structured sets of texts used for statistical analysis and hypothesis testing, checking occurrences or validating linguistic rules. By extension, system outputs depend heavily on the quality and the size of corpora. Large and reliable bilingual corpora are unavailable for many language pairs. In addition, translating into morphologically rich target languages makes the development of data-driven systems rather more difficult.
The event will highlight multisite research effort addressing approaches to integrating structural information into open-source solutions and the development of novel automatic machine translation evaluation using linguistically motivated metrics.For further information, please visit: http://ixa2.si.ehu.es/lihmt2011/(opens in new window)
Data-driven machine translation is increasingly used prevalent within the machine translation research community. From a quality point of view, results are becoming increasingly useful, especially when the target language is English.
Data-driven machine translation systems base their knowledge on bilingual aligned corpora, which are large and structured sets of texts used for statistical analysis and hypothesis testing, checking occurrences or validating linguistic rules. By extension, system outputs depend heavily on the quality and the size of corpora. Large and reliable bilingual corpora are unavailable for many language pairs. In addition, translating into morphologically rich target languages makes the development of data-driven systems rather more difficult.
The event will highlight multisite research effort addressing approaches to integrating structural information into open-source solutions and the development of novel automatic machine translation evaluation using linguistically motivated metrics.For further information, please visit: http://ixa2.si.ehu.es/lihmt2011/(opens in new window)