Integration of Multiple Bilingually-Trained Segmentation Schemes into Statistical Machine Translation

Michael PAUL  Andrew FINCH  Eiichiro SUMITA  

IEICE TRANSACTIONS on Information and Systems   Vol.E94-D   No.3   pp.690-697
Publication Date: 2011/03/01
Online ISSN: 1745-1361
DOI: 10.1587/transinf.E94.D.690
Print ISSN: 0916-8532
Type of Manuscript: PAPER
Category: Natural Language Processing
statistical machine translation,  word segmentation,  machine learning,  Asian languages,  

Full Text: PDF>>
Buy this Article

This paper proposes an unsupervised word segmentation algorithm that identifies word boundaries in continuous source language text in order to improve the translation quality of statistical machine translation (SMT) approaches. The method can be applied to any language pair in which the source language is unsegmented and the target language segmentation is known. In the first step, an iterative bootstrap method is applied to learn multiple segmentation schemes that are consistent with the phrasal segmentations of an SMT system trained on the resegmented bitext. In the second step, multiple segmentation schemes are integrated into a single SMT system by characterizing the source language side and merging identical translation pairs of differently segmented SMT models. Experimental results translating five Asian languages into English revealed that the proposed method of integrating multiple segmentation schemes outperforms SMT models trained on any of the learned word segmentations and performs comparably to available monolingually built segmentation tools.