Semi-Supervised Training for Statistical Word Alignment

We introduce a semi-supervised approach to training for statistical machine translation that alternates the traditional Expectation Maximization step that is applied on a large training corpus with a discriminative step aimed at increasing word-alignment quality on a small, manually word-aligned sub-corpus. We show that our algorithm leads not only to improved alignments but also to machine translation outputs of higher quality.

• Type: Conference
• Year: 2006
• Where: ACL
• Authors: Alexander Fraser, Daniel Marcu
• Keywords: ACL 2006, ACL 2007, Expectation Maximization Step, Machine Translation, Statistical Machine Translation