<?xml version="1.0" standalone="yes"?>
<Paper uid="E06-2014">
  <Title>ASSIST: Automated semantic assistance for translators</Title>
  <Section position="4" start_page="141" end_page="141" type="concl">
    <SectionTitle>
3 Conclusions
</SectionTitle>
    <Paragraph position="0"> The set of tools is now under further development.</Paragraph>
    <Paragraph position="1"> This involves an extension of the English semantic tagger, development of the Russian tagger with the target lexical coverage of 90% of source texts, designing the procedure for retrieval of semantically similar situations and completing the user interface. Identification of semantically similar situations can be improved by the use of segmentmatching algorithms as employed in Example-Based MT and translation memories (Planas and Furuse, 2000; Carl and Way, 2003).</Paragraph>
    <Paragraph position="2"> There are two main applications of the proposed methodology. One concerns training translators and advanced foreign language (FL) learners to make them aware of the variety of translation equivalents beyond the set offered by the dictionary. The other application pertains to the development of tools for practising translators. Although the Russian language is not typologically close to English and uses another writing system which does not allow easy identification of cognates, Russian and English belong to the same Indo-European family and the contents of Russian and English newspapers reflect the same set of topics. Nevertheless, the application of this research need not be restricted to the English-Russian pair only. The methodology for multilingual processing of monolingual comparable corpora, first tested in this project, will provide a blueprint for the development of similar tools for other language combinations.</Paragraph>
  </Section>
class="xml-element"></Paper>