<?xml version="1.0" standalone="yes"?>
<Paper uid="C04-1053">
  <Title>Evaluating Cross-Language Annotation Transfer in the MultiSemCor Corpus</Title>
  <Section position="8" start_page="1" end_page="1" type="concl">
    <SectionTitle>
6 Conclusion and future directions
</SectionTitle>
    <Paragraph position="0"> We have presented and evaluated an approach to the creation of high quality semantically annotated resources based on the exploitation of aligned parallel corpora. The results obtained from the thorough evaluation of the different steps involved in the methodology confirm the feasibility of the MultiSemCor project. The cross-lingual annotation transfer methodology is going to be applied also to the remaining 157 SemCor texts, which are currently being translated into Italian.</Paragraph>
    <Paragraph position="1"> As regards future research directions within the transfer annotation paradigm, it would be interesting to extend the methodology to other languages, e.g. Spanish, for which a WordNet exists and can be aligned with MultiWordNet.</Paragraph>
    <Paragraph position="2"> Moreover, as the Brown Corpus, used to create SemCor, has been syntactically annotated within the English Penn Treebank, the syntactic annotations of the SemCor texts are also available.</Paragraph>
    <Paragraph position="3"> We are planning to explore the possibility of transferring the syntactic annotations from the English to the Italian texts of MultiSemCor.</Paragraph>
  </Section>
class="xml-element"></Paper>