Improving statistical machine translation using bayesian word alignment and gibbs sampling
IEEE Transactions on Audio, Speech and Language Processing, cilt.21, sa.5, ss.1090-1101, 2013 (SCI-Expanded, Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 21 Sayı: 5
- Basım Tarihi: 2013
- Doi Numarası: 10.1109/tasl.2013.2244087
- Dergi Adı: IEEE Transactions on Audio, Speech and Language Processing
- Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
- Sayfa Sayıları: ss.1090-1101
- Anahtar Kelimeler: Bayesian methods, Gibbs sampling, statistical machine translation (SMT), word alignment
- Boğaziçi Üniversitesi Adresli: Evet
Özet
We present a Bayesian approach to word alignment inference in IBM Models 1 and 2. In the original approach, word translation probabilities (i.e., model parameters) are estimated using the expectation-maximization (EM) algorithm. In the proposed approach, they are random variables with a prior and are integrated out during inference. We use Gibbs sampling to infer the word alignment posteriors. The inferred word alignments are compared against EM and variational Bayes (VB) inference in terms of their end-to-end translation performance on several language pairs and types of corpora up to 15 million sentence pairs. We show that Bayesian inference outperforms both EM and VB in the majority of test cases. Further analysis reveals that the proposed method effectively addresses the high-fertility rare word problem in EM and unaligned rare word problem in VB, achieves higher agreement and vocabulary coverage rates than both, and leads to smaller phrase tables. © 2006-2012 IEEE.