As a linguistic phenomenon, collocations have been the subject of numerous re- searches both in the fields of theoreti- cal and descriptive linguistics, and, more recently, in automatic Natural Language Processing. In the area of Machine there is still improvements to be done, as ma- jor translation engines do not handle col- locations in the appropriate way and end up producing literal unsatisfactory transla- tions. Having as a starting point our pre- vious work on machine translation error analysis (Costa et al., 2015), in this ar- ticle we present a corpus annotated with collocation errors and their classification. To our believe, to have a clear understand- ing of the difficulties that the collocations represent to the Machine Translations en- gines, it is necessary a detailed linguistic analysis of their errors.
|Title of host publication||Workshop Proceedings on Multi-word Units in Machine Translation and Translation Technology|
|Editors||Gloria Pastor, Johanna Monti, Violeta Seretan, Ruslan Mitkov|
|Place of Publication||Geneva|
|Number of pages||4|
|Publication status||Published - 2016|
|Event||EUROPHRAS 2015 - Malaga, Spain|
Duration: 29 Jun 2015 → 2 Jul 2015
|Period||29/06/15 → 2/07/15|
- Machine Translation
Costa, Â., Lino, M. T. R. F., & Coheur, L. (2016). Challenges on the Automatic Translation of Collocations. In G. Pastor, J. Monti, V. Seretan, & R. Mitkov (Eds.), Workshop Proceedings on Multi-word Units in Machine Translation and Translation Technology (pp. 74-77). Editions Tradulex.