<?xml version="1.0" standalone="yes"?>
<Paper uid="C04-1045">
  <Title>Improving Word Alignment Quality using Morpho-syntactic Information</Title>
  <Section position="3" start_page="0" end_page="0" type="relat">
    <SectionTitle>
2 Related Work
</SectionTitle>
    <Paragraph position="0"> The popular IBM models for statistical machine translation are described in (Brown et al., 1993) and the HMM-based alignment model was introduced in (Vogel et al., 1996). A good overview of all these models is given in (Och and Ney, 2003) where the model IBM-6 is also introduced as the log-linear interpolation of the other models.</Paragraph>
    <Paragraph position="1"> Context dependencies have been introduced into the training of alignments in (Varea et al., 2002), but they do not take any linguistic information into account.</Paragraph>
    <Paragraph position="2"> Some recent publications have proposed the use of morpho-syntactic knowledge for statistical machine translation, but mostly only for the preprocessing step whereas training procedure of the statistical models remains the same (e.g. (Niessen and Ney, 2001a)).</Paragraph>
    <Paragraph position="3"> Incorporation of the morpho-syntactic knowlegde into statistical models has been dealt in (Niessen and Ney, 2001): hierarchical lexicon models containing base forms and set of morpho-syntactic tags are proposed for the translation from German into English. However, these lexicon models are not used for the training but have been created from the Viterbi alignment obtained after the usual training procedure. null The use of POS information for improving statistical alignment quality of the HMM-based model is described in (Toutanova et al., 2002). They introduce additional lexicon probability for POS tags in both languages, but actually are not going beyond full forms.</Paragraph>
  </Section>
class="xml-element"></Paper>