论文信息 - Symmetric Probabilistic Alignment

Symmetric Probabilistic Alignment

We recently decided to develop a new alignment algorithm for the purpose of improving our Example-Based Machine Translation (EBMT) system's performance, since subsentential alignment is critical in locating the correct translation for a matched fragment of the input. Unlike most algorithms in the literature, this new Symmetric Probabilistic Alignment (SPA) algorithm treats the source and target languages in a symmetric fashion. In this short paper, we outline our basic algorithm and some extensions for using context and positional information, and compare its alignment accuracy on the Romanian-English data for the shared task with IBM Model 4 and the reported results from the prior workshop.

Jaime G. Carbonell | Ralf D. Brown | Jae Dong Kim | Peter J. Jansen

[1] Sergei Nirenburg,et al. Integrating Translations from Multiple Sources within the PANGLOSS Mark III Machine Translation System , 1994, AMTA.

[2] Ralf D. Brown,et al. Automated Generalization of Translation Examples , 2000, COLING.

[3] Daniel Marcu,et al. A Phrase-Based, Joint Probability for Statistical Machine Translation , 2002 .

[4] Ralf D. Brown. Automated Dictionary Extraction for “Knowledge-Free” Example-Based Translation , 2006 .

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

[6] Paul N. Bennett,et al. Reducing boundary friction using translation-fragment overlap , 2003, MTSUMMIT.

[7] Ted Pedersen,et al. An Evaluation Exercise for Word Alignment , 2003, ParallelTexts@NAACL-HLT.

[8] Jaime G. Carbonell,et al. Symmetric probabilistic alignment for example-based translation , 2005, EAMT.

[9] Daniel Marcu,et al. A Phrase-Based,Joint Probability Model for Statistical Machine Translation , 2002, EMNLP.

[10] Salim Roukos,et al. Bleu: a Method for Automatic Evaluation of Machine Translation , 2002, ACL.