Applying morphological decomposition to statistical machine translation

This paper describes the Aalto submission for the German-to-English and the Czechto-English translation tasks of the ACL 2010 Joint Fifth Workshop on Statistical Machine Translation and MetricsMATR. Statistical machine translation has focused on using words, and longer phrases constructed from words, as tokens in the system. In contrast, we apply different morphological decompositions of words using the unsupervised Morfessor algorithms. While translation models trained using the morphological decompositions did not improve the BLEU scores, we show that the Minimum Bayes Risk combination with a word-based translation model produces significant improvements for the Germanto-English translation. However, we did not see improvements for the Czech-toEnglish translations.