<?xml version="1.0" standalone="yes"?>
<Paper uid="P93-1004">
  <Title>STRUCTURAL MATCHING OF PARALLEL TEXTS</Title>
  <Section position="12" start_page="29" end_page="29" type="concl">
    <SectionTitle>
CONCLUSIONS
</SectionTitle>
    <Paragraph position="0"> The structural matchillg problem of parallel texts is formally defined and our current implementation and experilnents are introduced. Although the research is at the preliminary stage and has a. very simple setting, the experiments have shown a. nulnber of interesting results. The method is easily enhanced by ilnproving the gramnm.rs and by incorporating more accurate similarity measurement.</Paragraph>
    <Paragraph position="1"> Number of other researches of building tra.nslation dictionaries and of deterlnining similarity relationship between words are useful to improve our method.</Paragraph>
    <Paragraph position="2"> To extract useful information fl'om bilingual corpora, structural matching is inevitable for language pairs like English and Japanese that have quite different linguistic structure. Incidentally, we have found that this dissimilarity plays an important role in resolving syntactic ambiguities since the sources of anlbiguities in English and Japanese sentences are in many cases do not coincide (Utsuro 92). We are currently working on extracting verbal case frames of Japanese fi'om the results of structural matching of a aal)anese-l~nglish corpus (Utsuro 93). The salne teclmique is naturally a.pplicable to acquire verbal case fi'ames of English as well. Another application we are envisaging is to extract translation pattern from the results of structural matching.</Paragraph>
    <Paragraph position="3"> We plan to work on possible improvements discussed in the preceding section, and will make large scale experiments using translated newspal~er articles, based on the phrase matching stra.t.egy.</Paragraph>
  </Section>
class="xml-element"></Paper>