DUSTer: A Method for Unraveling Cross-Language Divergences for Statistical Word-Level Alignment

Abstract

The frequent occurrence of divergences|structural diier-ences between languages|presents a great challenge for statistical word-level alignment. In this paper, we introduce DUSTer, a method for systematically identifying common divergence types and transforming an English sentence structure to bear a closer resemblance to that of another language. Our ultimate goal is to enable more accurate alignment and projection of dependency trees in another language without requiring any training on dependency-tree data in that language. We present an empirical analysis comparing the complexities of performing word-level alignments with and without divergence handling. Our results suggest that our approach facilitates word-level alignment, particularly for sentence pairs containing divergences.

DOI: 10.1007/3-540-45820-4_4

Extracted Key Phrases

Cite this paper

@inproceedings{Dorr2002DUSTerAM, title={DUSTer: A Method for Unraveling Cross-Language Divergences for Statistical Word-Level Alignment}, author={Bonnie J. Dorr and Lisa Pearl and Rebecca Hwa and Nizar Habash}, booktitle={AMTA}, year={2002} }