For Full-Text PDF, please login, if you are a member of IEICE,|
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
Constraining a Generative Word Alignment Model with Discriminative Output
Chooi-Ling GOH Taro WATANABE Hirofumi YAMAMOTO Eiichiro SUMITA
IEICE TRANSACTIONS on Information and Systems
Publication Date: 2010/07/01
Online ISSN: 1745-1361
Print ISSN: 0916-8532
Type of Manuscript: PAPER
Category: Natural Language Processing
word alignment, discriminative model, generative model, hybrid, SMT,
Full Text: PDF(646.5KB)>>
We present a method to constrain a statistical generative word alignment model with the output from a discriminative model. The discriminative model is trained using a small set of hand-aligned data that ensures higher precision in alignment. On the other hand, the generative model improves the recall of alignment. By combining these two models, the alignment output becomes more suitable for use in developing a translation model for a phrase-based statistical machine translation (SMT) system. Our experimental results show that the joint alignment model improves the translation performance. The improvement in average of BLEU and METEOR scores is around 1.0-3.9 points.