<?xml version="1.0" standalone="yes"?>
<Paper uid="C02-1099">
  <Title>An English-Korean Transliteration Model Using Pronunciation and Contextual Rules</Title>
  <Section position="3" start_page="11" end_page="11" type="concl">
    <SectionTitle>
5. Conclusion
</SectionTitle>
    <Paragraph position="0"> We propose an English-Korean transliteration model using pronunciation and contextual rules.</Paragraph>
    <Paragraph position="1"> Unlike the previous works, our method use phonetic and orthographic information for transliteration. With them our method showed significant performance increase about 31%. We also showed that origin of English words was important in E-K transliteration.</Paragraph>
    <Paragraph position="2"> In future works, a study is attempting to develop a method for handling English of various foreign origin, which this paper did not handle. To improve accuracy, contextual rules must be added using larger data. Our method may be useful to many NLP applications such as EPU may produce many phonemes, it may be difficult to acquire a good result without context of phoneme and EPU.  English word (correct transliteration / transliteration by the previous works) automatic bi-lingual dictionary construction, information retrieval, machine translation, speech recognition and so on.</Paragraph>
  </Section>
class="xml-element"></Paper>