论文信息 - Word association models and search strategies for discriminative word alignment

Word association models and search strategies for discriminative word alignment

This paper deals with core aspects of discriminative word alignment systems, namely basic word association models as well as search strategies. We compare various low-computational-cost word as- sociation models: 2 score, log-likelihood ratio and IBM model 1. We also compare three beam-search strategies. We show that it is more ex- ible and accurate to let links to the same word compete together, than introducing them sequentially in the alignment hypotheses, which is the strategy followed in several systems.

Rafael E. Banchs | Patrik Lambert

[1] Kenneth Ward Church,et al. Identifying Word Correspondences in Parallel Texts , 1991, HLT.

[2] Robert C. Moore. A Discriminative Framework for Bilingual Word Alignment , 2005, HLT.

[3] Colin Cherry,et al. A Probability Model to Improve Word Alignment , 2003, ACL.

[4] Yang Liu,et al. Log-Linear Models for Word Alignment , 2005, ACL.

[5] Hermann Ney,et al. A Systematic Comparison of Various Statistical Alignment Models , 2003, CL.

[6] Robert L. Mercer,et al. The Mathematics of Statistical Machine Translation: Parameter Estimation , 1993, CL.

[7] Rafael E. Banchs,et al. Discriminative Alignment Training without Annotated Data for Machine Translation , 2007, HLT-NAACL.

[8] Robert C. Moore. On Log-Likelihood-Ratios and the Significance of Rare Events , 2004, EMNLP.

[9] I. Dan Melamed,et al. Models of translation equivalence among words , 2000, CL.

[10] Ted Dunning,et al. Accurate Methods for the Statistics of Surprise and Coincidence , 1993, CL.

[11] José B. Mariño,et al. Guidelines for Word Alignment Evaluation and Manual Alignment , 2005, Lang. Resour. Evaluation.

[12] Kenneth Ward Church,et al. Identifying word correspondence in parallel texts , 1991 .