Phrase Based Language Model for Statistical Machine Translation: Empirical Study
This work addresses reordering issues in machine translation systems, but it is incremental as it builds on existing phrase-based approaches from speech recognition.
The authors tackled the challenge of reordering in machine translation by proposing two phrase-based language models that use phrases as constituent units, which outperformed word-based models in perplexity and n-best list re-ranking.
Reordering is a challenge to machine translation (MT) systems. In MT, the widely used approach is to apply word based language model (LM) which considers the constituent units of a sentence as words. In speech recognition (SR), some phrase based LM have been proposed. However, those LMs are not necessarily suitable or optimal for reordering. We propose two phrase based LMs which considers the constituent units of a sentence as phrases. Experiments show that our phrase based LMs outperform the word based LM with the respect of perplexity and n-best list re-ranking.