Hebrew Morphological Preprocessing for Statistical Machine Translation
Nimesh Singh, Nizar Y. Habash · 2012
This paper presents a range of preprocessing solutions for Hebrew-English statistical machine translation. Our best system, using a morphological analyzer, increases 3.5 BLEU points over a no-tokenization baseline on a blind test set. The next best system uses Morfessor, an unsupervised morphological segmenter, and obtains almost 3.0 BLEU points over the baseline. 1