<?xml version="1.0" standalone="yes"?>
<Paper uid="W06-3108">
  <Title>Discriminative Reordering Models for Statistical Machine Translation</Title>
  <Section position="2" start_page="0" end_page="0" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> In recent evaluations, phrase-based statistical machine translation systems have achieved good performance. Still the fluency of the machine translation output leaves much to desire. One reason is that most phrase-based systems use a very simple re-ordering model. Usually, the costs for phrase movements are linear in the distance, e.g. see (Och et al., 1999; Koehn, 2004; Zens et al., 2005).</Paragraph>
    <Paragraph position="1"> Recently, in (Tillmann and Zhang, 2005) and in (Koehn et al., 2005), a reordering model has been described that tries to predict the orientation of a phrase, i.e. it answers the question 'should the next phrase be to the left or to the right of the current phrase?' This phrase orientation probability is conditioned on the current source and target phrase and relative frequencies are used to estimate the probabilities. null We adopt the idea of predicting the orientation, but we propose to use a maximum-entropy based model. The relative-frequency based approach may suffer from the data sparseness problem, because most of the phrases occur only once in the training corpus. Our approach circumvents this problem by using a combination of phrase-level and word-level features and by using word-classes or part-of-speech information. Maximum entropy is a suitable framework for combining these different features with a well-defined training criterion.</Paragraph>
    <Paragraph position="2"> In (Koehn et al., 2005) several variants of the orientation model have been tried. It turned out that for different tasks, different models show the best performance. Here, we let the maximum entropy training decide which features are important and which features can be neglected. We will see that additional features do not hurt performance and can be safely added to the model.</Paragraph>
    <Paragraph position="3"> The remaining part is structured as follows: first we will describe the related work in Section 2 and give a brief description of the baseline system in Section 3. Then, we will present the discriminative reordering model in Section 4. Afterwards, we will evaluate the performance of this new model in Section 5. This evaluation consists of two parts: first we will evaluate the prediction capabilities of the model on a word-aligned corpus and second we will show improved translation quality compared to the base-line system. Finally, we will conclude in Section 6.</Paragraph>
  </Section>
class="xml-element"></Paper>