Paper: Modeling with Structures in Statistical Machine Translation

ACL ID C98-2216
Title Modeling with Structures in Statistical Machine Translation
Venue International Conference on Computational Linguistics
Session Main Conference
Year 1998
Authors

Most statistical machine translation systems employ a word-based alignment model. In this paper we demonstrate that word-based align: ment is a major cause of translation errors. We propose a new alignment model based on shal- low phrase structures, and tile structures can be automatically acquired from parallel corpus. This new model achieved over 110% error reduc- tion for our st)oken language translation task.