Normalizing German and English Inflectional Morphology to Improve Statistical Word Alignment
- Simon Corston-Oliver ,
- Michael Gamon
Published by Association for Machine Translation in the Americas
German has a richer system of inflectional morphology than English, which causes problems for current approaches to statistical word alignment. Using Giza++ as a reference implementation of the IBM Model 1, an HMM-based alignment and IBM Model 4, we measure the impact of normalizing inflectional morphology on German-English statistical word alignment. We demonstrate that normalizing inflectional morphology improves the perplexity of models and reduces alignment errors.