<?xml version="1.0" standalone="yes"?>
<Paper uid="W03-1110">
  <Title>Issues in Preand Post-translation Document Expansion: Untranslatable Cognates and Missegmented Words</Title>
  <Section position="3" start_page="0" end_page="0" type="intro">
    <SectionTitle>
2 Related Work
</SectionTitle>
    <Paragraph position="0"> This work draws on prior research in pseudo-relevance feedback for both queries and documents.</Paragraph>
    <Section position="1" start_page="0" end_page="0" type="sub_section">
      <SectionTitle>
2.1 Pre- and Post-translation Query Expansion
</SectionTitle>
      <Paragraph position="0"> In pre-translation query expansion, the goal is both that of monolingual query expansion - providing additional terms to refine the query and to enhance the probability of matching the terminology chosen by the authors of the document - and to provide additional terms to limit the possibility of failing to translate a concept in the query simply because the particular term is not present in the translation lexicon. (Ballesteros and Croft, 1997) evaluated pre- and post-translation query expansion in a Spanish-English cross-language information retrieval task and found that combining pre- and post-translation query expansion improved both precision and recall with pre-translation expansion improving both precision and recall, and post-translation expansion enhancing precision. (McNamee and Mayfield, 2002)'s dictionary ablation experiments on the effect of translation resource size and pre- and post-translation query expansion effectiveness demonstrated the key and dominant role of pre-translation expansion in providing translatable terms. If too few terms are translated, post-translation expansion can provide little improvement.</Paragraph>
    </Section>
    <Section position="2" start_page="0" end_page="0" type="sub_section">
      <SectionTitle>
2.2 Document Expansion
</SectionTitle>
      <Paragraph position="0"> The document expansion approach was first proposed by (Singhal et al., 1999) in the context of spoken document retrieval. Since spoken document retrieval involves search of error-prone automatic speech recognition transcriptions, Singhal et al introduced document expansion as a way of recovering those words that might have been in the original broadcast but that had been misrecognized. They speculated that correctly recognized terms would yield a topically coherent transcript, while the sporadic errors would be from a random distribution.</Paragraph>
      <Paragraph position="1"> Enriching the documents with highly selective terms drawn from highly ranked documents retrieved by using the document itself as a query yielded retrieval effectiveness that improved not only over the original errorful transcription but also over a perfect manual transcription. (Levow and Oard, 2000) applied post-translation document expansion to both spoken documents and newswire text in Mandarin-English multi-lingual retrieval and found some improvements in retrieval effectiveness. (Levow, 2003) evaluated multi-scale units (words and bigrams) for post-transcription expansion of Mandarin spoken documents, finding the significant improvements for expansion with word units using bigram based indexing.</Paragraph>
    </Section>
  </Section>
class="xml-element"></Paper>