We propose a novel language-independent approach for improving machine translation for resource-poor languages by exploiting their similarity to resource-rich ones. More precisely, we improve the translation from a resource-poor source language X1 into a resourcerich language Y given a bi-text containing a limited number of parallel sentences for X 1-Y and a larger bi-text for X 2-Y for some resource-rich language X 2 that is closely related to X 1. This is achieved by taking advantage of the opportunities that vocabulary overlap and similarities between the languages X 1 and X 2 in spelling, word order, and syntax offer: (1) we improve the word alignments for the resource-poor language, (2) we further augment it with additional translation options, and (3) we take care of potential spelling differences through appropriate transliteration. The evaluation for Indonesian!English using Malay and for Spanish!English using Portuguese and pretending Spanish is resource-poor shows an absolute gain of up to 1.35 and 3.37 BLEU points, respectively, which is an improvement over the best rivaling approaches, while using much less additional data. Overall, our method cuts the amount of necessary "real" training data by a factor of 2-5.
|Number of pages||44|
|Journal||Journal of Artificial Intelligence Research|
|Publication status||Published - 1 May 2012|
ASJC Scopus subject areas
- Artificial Intelligence