Paper: Learning Better Rule Extraction with Translation Span Alignment

ACL ID P12-2055
Title Learning Better Rule Extraction with Translation Span Alignment
Venue Annual Meeting of the Association of Computational Linguistics
Session Short Paper
Year 2012
Authors

This paper presents an unsupervised ap- proach to learning translation span align- ments from parallel data that improves syntactic rule extraction by deleting spuri- ous word alignment links and adding new valuable links based on bilingual transla- tion span correspondences. Experiments on Chinese-English translation demonstrate improvements over standard methods for tree-to-string and tree-to-tree translation.