<?xml version="1.0" standalone="yes"?>
<Paper uid="C02-1016">
  <Title>Determining Recurrent Sound Correspondences by Inducing Translation Models</Title>
  <Section position="7" start_page="2" end_page="2" type="concl">
    <SectionTitle>
7 Conclusions and future work
</SectionTitle>
    <Paragraph position="0"> I have presented a novel approach to the determination of correspondences in bilingual wordlists.</Paragraph>
    <Paragraph position="1"> The results of experiments indicate that the approach is robust enough to handle a substantial amount of noise that is introduced by unrelated word pairs. CORDI does well even when the number of non-cognate pairs is more than double the number of cognate pairs. When tested on the cognate-identification task, CORDI achieves substantially higher precision than comparable programs. The correspondences are explicitly posited, which means that, unlike in some statistical approaches, they can be verified by examining individual cognate pairs. In contrast with approaches that assume a rigid alignment based on the syllabic structure, the models presented here can link phonemes in any word position.</Paragraph>
    <Paragraph position="2"> Currently, I am working on the incorporation of complex correspondences into the cognate identification algorithm by employing Melamed's (1997) algorithm for discovering non-compositional compounds in parallel data. Such an extension would overcome the limitation of the one-to-one model, in which links are induced only between individual phonemes. Other possible extensions include taking into account the phonological context of correspondences, combining the correspondence-based approach with phonetic-based approaches, and identifying correspondences and cognates directly in dictionary-type data.</Paragraph>
    <Paragraph position="3"> The results presented here prove that the techniques developed in the context of statistical machine translation can be successfully applied to a problem in diachronic phonology. The transfer of methods and insights should also be possible in the other direction.</Paragraph>
  </Section>
class="xml-element"></Paper>