Paper: Translation Model Pruning via Usage Statistics for Statistical Machine Translation

ACL ID N07-2006
Title Translation Model Pruning via Usage Statistics for Statistical Machine Translation
Venue Human Language Technologies
Session Short Paper
Year 2007
Authors

We describe a new pruning approach to remove phrase pairs from translation mod- els of statistical machine translation sys- tems. The approach applies the original translation system to a large amount of text and calculates usage statistics for the phrase pairs. Using these statistics the rele- vance of each phrase pair can be estimated. The approach is tested against a strong baseline based on previous work and shows significant improvements.