Decision Trees for Lexical Smoothing in Statistical Machine Translation

We present a method for incorporating arbitrary context-informed word attributes into statistical machine translation by clustering attribute-quali ed source words, and smoothing their word translation probabilities using binary decision trees. We describe two ways in which the decision trees are used in machine translation: by using the attribute-quali ed source word clusters directly, or by using attributedependent lexical translation probabilities that are obtained from the trees, as a lexical smoothing feature in the decoder model. We present experiments using Arabic-to-English newswire data, and using Arabic diacritics and part-ofspeech as source word attributes, and show that the proposed method improves on a state-of-the-art translation