Paper: Selective Phrase Pair Extraction for Improved Statistical Machine Translation

ACL ID N07-2053
Title Selective Phrase Pair Extraction for Improved Statistical Machine Translation
Venue Human Language Technologies
Session Short Paper
Year 2007
Authors

Phrase-based statistical machine transla- tionsystemsdependheavilyontheknowl- edge represented in their phrase transla- tion tables. However, the phrase pairs included in these tables are typically se- lected using simple heuristics that poten- tially leave much room for improvement. In this paper, we present a technique for selecting the phrase pairs to include in phrase translation tables based on their es- timated quality according to a translation model. This method not only reduces the size of the phrase translation table, but also improves translation quality as mea- sured by the BLEU metric.