<?xml version="1.0" standalone="yes"?>
<Paper uid="W06-3123">
  <Title>Constraining the Phrase-Based, Joint Probability Statistical Translation Model</Title>
  <Section position="2" start_page="0" end_page="154" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> Machine translation is a hard problem because of the highly complex, irregular and diverse nature of natural languages. It is impossible to accurately model all the linguistic rules that shape the translation process, and therefore a principled approach uses statistical methods to make optimal decisions given incomplete data.</Paragraph>
    <Paragraph position="1"> The original IBM Models (Brown et al., 1993) learn word-to-word alignment probabilities which makes it computationally feasible to estimate model parameters from large amounts of training data. Phrase-based SMT models, such as the alignment template model (Och, 2003), improve on word-based models because phrases provide local context which leads to better lexical choice and more reliable local reordering. However, most phrase-based models extract their phrase pairs from previously word-aligned corpora using ad-hoc heuristics. These models perform no search for optimal phrasal alignments. Even though this is an efficient strategy, it is a departure from the rigorous statistical framework of the IBM Models.</Paragraph>
    <Paragraph position="2"> Marcu and Wong (2002) proposed the joint probability model which directly estimates the phrase translation probabilities from the corpus in a theoretically governed way. This model neither relies on potentially sub-optimal word alignments nor on heuristics for phrase extraction. Instead, it searches the phrasal alignment space, simultaneously learning translation lexicons for both words and phrases. The joint model has been shown to outperform standard models on restricted data sets such as the small data track for Chinese-English in the 2004 NIST MT Evaluation (Przybocki, 2004).</Paragraph>
    <Paragraph position="3"> However, considering all possible phrases and all their possible alignments vastly increases the computational complexity of the joint model when compared to its word-based counterpart. In this paper, we propose a method of constraining the search space of the joint model to areas where most of the unpromising phrasal alignments are eliminated and yet as many potentially useful alignments as possible are still explored. The joint model is constrained to phrasal alignments whichdonotcontradictasethighconfidenceword alignments for each sentence. These high confidence alignments could incorporate information from both statistical and linguistic sources. In this paper we use the points of high confidence from the intersection of the bi-directional Viterbi word alignments to constrain the model, increasing performance and decreasing complexity.</Paragraph>
  </Section>
class="xml-element"></Paper>