Domain Adaptation in Statistical Machine Translation using Factored Translation Models

In recent years the performance of SMT increased in domains with enough training data. But under real-world conditions, it is often not possible to collect enough parallel data. We propose an approach to adapt an SMT system using small amounts of parallel in-domain data by introducing the corpus identifier (corpus id) as an additional target factor. Then we added features to model the generation of the tags and features to judge a sequence of tags. Using this approach we could improve the translation performance in two domains by up to 1 BLEU point when translating from German to English.