<?xml version="1.0" standalone="yes"?>
<Paper uid="P06-1122">
  <Title>Modelling lexical redundancy for machine translation</Title>
  <Section position="9" start_page="975" end_page="975" type="concl">
    <SectionTitle>
7 Conclusions and Future Work
</SectionTitle>
    <Paragraph position="0"> We proposed a framework for modelling lexical redundancy in machine translation and tackled optimisation of the lexicon via Bayesian model selection over a set of cluster-based translation models. We showed improvements in translation quality incorporating these models within a phrase-based SMT sytem. Additional gains resulted from the inclusion of an MRF prior over model structure. We demonstrated that this prior could be usedtolearnweightsformonolingualfeaturesthat characterise bilingual redundancy. Preliminary experiments defining MRF features over morphological annotation suggest this model can also identify redundant distinctions categorised linguistically (for instance, that morphological case is redundant on Czech nouns and adjectives with respecttoEnglish,whilenumber isredundantonly on adjectives). In future work we will investigate theuseoflinguisticresourcestodefinefeaturesets for the MRF prior. Lexical redundancy would ideally be addressed in the context of phrases, however, computation and statistical estimation may then be significantly more challenging.</Paragraph>
  </Section>
class="xml-element"></Paper>