The usefulness of parallel corpora in translation studies and machine translation is strictly related to the availability of aligned data. In this paper we discuss the issues related to the design of a tool for the alignment of data from a parallel treebank, which takes into account morphological, syntactic and semantic knowledge as annotated in this kind of resource. A preliminary analysis is presented which is based on a case study, a parallel treebank for Italian, English and French, i.e. ParTUT. The paper will focus, in particular, on the study of translational divergences and their implications for the development of an alignment tool of parallel parse trees that, benefitting from the linguistic information provided in ParTUT, could properly deal with such divergences.
Translational Divergences and Their Alignment in a Parallel Multilingual Treebank
Sanguinetti Manuela;
2012-01-01
Abstract
The usefulness of parallel corpora in translation studies and machine translation is strictly related to the availability of aligned data. In this paper we discuss the issues related to the design of a tool for the alignment of data from a parallel treebank, which takes into account morphological, syntactic and semantic knowledge as annotated in this kind of resource. A preliminary analysis is presented which is based on a case study, a parallel treebank for Italian, English and French, i.e. ParTUT. The paper will focus, in particular, on the study of translational divergences and their implications for the development of an alignment tool of parallel parse trees that, benefitting from the linguistic information provided in ParTUT, could properly deal with such divergences.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
