论文信息 - Sub-phrasal matching and structural templates in example-based MT

Sub-phrasal matching and structural templates in example-based MT

In this work I look at two different paradigms of Example-Based Machine Translation (EBMT). I combine the strengths of these two systems and build a new EBMT engine that combines sub-phrasal matching with structural templates. This synthesis results in higher translation quality and more graceful degradation, yielding 1.5% to 7.5% relative improvement in BLEU scores.

Aaron B. Phillips

[1] Ralf D. Brown,et al. Example-Based Machine Translation in the Pangloss System , 1996, COLING.

[2] Adwait Ratnaparkhi,et al. A Maximum Entropy Model for Part-Of-Speech Tagging , 1996, EMNLP.

[3] Alon Lavie,et al. Automatic rule learning for resource-limited MT , 2002, AMTA.

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

[5] Daniel Jurafsky,et al. Automatic Tagging of Arabic Text: From Raw Text to Base Phrase Chunks , 2004, NAACL.

[6] Ralf D. Brown. A Modified Burrows-Wheeler Transform for Highly Scalable Example-Based Translation , 2004, AMTA.

[7] S. Vogel. PESA: Phrase Pair Extraction as Sentence Splitting , 2005, MTSUMMIT.

[8] Ying Zhang,et al. An efficient phrase-to-phrase alignment model for arbitrarily long phrase and large corpora , 2005, EAMT.

[9] Chris Callison-Burch,et al. Scaling Phrase-Based Statistical Machine Translation to Larger Corpora and Longer Phrases , 2005, ACL.

[10] Alon Lavie,et al. MT for Minority Languages Using Elicitation-Based Learning of Syntactic Transfer Rules , 2002, Machine Translation.