Improved phrase-based smt with syntactic reordering patterns learned from lattice scoring

In this paper, we present a novel approach to incorporate source-side syntactic reorder-ing patterns into phrase-based SMT. The main contribution of this work is to use the lat-tice scoring approach to exploit and utilize re-ordering information that is favoured by the baseline PBSMT system. By referring to the parse trees of the training corpus, we repre-sent the observed reorderings with source-side syntactic patterns. The extracted patterns are then used to convert the parsed inputs into word lattices, which contain both the origi-nal source sentences and their potential re-orderings. Weights of the word lattices are estimated from the observations of the syn-tactic reordering patterns in the training cor-pus. Finally, the PBSMT system is tuned and tested on the generated word lattices to show the benefits of adding potential source-side reorderings in the inputs. We confirmed the effectiveness of our proposed method on a medium-sized corpus for Chinese-English machine translation task. Our method out-performed the baseline system by 1.67 % rela-tive on a randomly selected testset and 8.56% relative on the NIST 2008 testset in terms of BLEU score.