Using Feature Structures to Improve Verb Translation in English-to-German Statistical MT
Philip J. Williams, Philipp Koehn · 2014
SCFG-based statistical MT models have proven effective for modelling syntactic aspects of translation, but still suffer prob-lems of overgeneration. The production of German verbal complexes is particu-larly challenging since highly discontigu-ous constructions must be formed con-sistently, often from multiple independent rules. We extend a strong SCFG-based string-to-tree model to incorporate a rich feature-structure based representation of German verbal complex types and com-pare verbal complex production against that of the reference translations, finding a high baseline rate of error. By developing model features that use source-side infor-mation to influence the production of ver-bal complexes we are able to substantially improve the type accuracy as compared to the reference. 1