<?xml version="1.0" standalone="yes"?>
<Paper uid="P03-1019">
  <Title>A Comparative Study on Reordering Constraints in Statistical Machine Translation</Title>
  <Section position="2" start_page="0" end_page="0" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> In statistical machine translation, we are given a source language ('French') sentence fJ1 = f1 :::fj :::fJ, which is to be translated into a target language ('English') sentence eI1 = e1 :::ei :::eI: Among all possible target language sentences, we will choose the sentence with the highest probability: null</Paragraph>
    <Paragraph position="2"> The decomposition into two knowledge sources in Eq. 2 is the so-called source-channel approach to statistical machine translation (Brown et al., 1990). It allows an independent modeling of target language model Pr(eI1) and translation model Pr(fJ1 jeI1). The target language model describes the well-formedness of the target language sentence.</Paragraph>
    <Paragraph position="3"> The translation model links the source language sentence to the target language sentence. It can be further decomposed into alignment and lexicon model.</Paragraph>
    <Paragraph position="4"> The argmax operation denotes the search problem, i.e. the generation of the output sentence in the target language. We have to maximize over all possible target language sentences.</Paragraph>
    <Paragraph position="5"> In this paper, we will focus on the alignment problem, i.e. the mapping between source sentence positions and target sentence positions. As the word order in source and target language may differ, the search algorithm has to allow certain word-reorderings. If arbitrary word-reorderings are allowed, the search problem is NP-hard (Knight, 1999). Therefore, we have to restrict the possible reorderings in some way to make the search problem feasible. Here, we will discuss two such constraints in detail. The first constraints are based on inversion transduction grammars (ITG) (Wu, 1995; Wu, 1997). In the following, we will call these the ITG constraints. The second constraints are the IBM constraints (Berger et al., 1996). In the next section, we will describe these constraints from a theoretical point of view. Then, we will describe the resulting search algorithm and its extension for word graph generation. Afterwards, we will analyze the Viterbi alignments produced during the training of the alignment models. Then, we will compare the translation results when restricting the search to either of these constraints.</Paragraph>
  </Section>
class="xml-element"></Paper>