Translation Model Based Weighting for Phrase Extraction

Domain adaptation for statistical machine translation is the task of altering general models to improve performance on the test domain. In this work, we suggest several novel weighting schemes based on translation models for adapted phrase extraction. To calculate the weights, we first phrase align the general bilingual training data, then, using domain specific translation models, the aligned data is scored and weights are defined over these scores. Experiments are performed on two translation tasks, German-to-English and Arabic-toEnglish translation with lectures as the target domain. Different weighting schemes based on translation models are compared, and significant improvements over automatic translation quality are reported. In addition, we compare our work to previous methods for adaptation and show significant gains.