<?xml version="1.0" standalone="yes"?>
<Paper uid="C02-2007">
  <Title>Building a Bilingual WordNet-Like Lexicon: the New Approach and Algorithms</Title>
  <Section position="5" start_page="2" end_page="2" type="concl">
    <SectionTitle>
4 Conclusions
</SectionTitle>
    <Paragraph position="0"> ICL, Peking University has launched the Project CCD since Sept., 2000. Due to the nice features of the new approach, we do have benefited a lot by employing it to build CCD.</Paragraph>
    <Paragraph position="1"> By now, we have fulfilled more than 32,000 Chinese-English concept pairs in noun.</Paragraph>
    <Paragraph position="2"> In the near future, ICL wants to come to a total amount of 100,000 or so bilingual concepts, which might largely meet our need of applications.</Paragraph>
    <Paragraph position="3"> What is more, as the byproducts of the new approach and experiences, we have even found some errors and faults of semantic expressing with WordNet 1.6.</Paragraph>
    <Paragraph position="4"> For example, in the lexicon there are many occurrences of a node with multiple-father in the identical category (772 times in noun, e.g. {radish}) or a node with single-father in the other category (2,172 times in noun, e.g.</Paragraph>
    <Paragraph position="5"> {prayer_wheel}).</Paragraph>
    <Paragraph position="6"> In verb, there even exists a node with father being oneself (e.g. {reserve, hold, book}).</Paragraph>
    <Paragraph position="7"> These phenomena are quite abnormal and puzzling according to the specification of WordNet. Something may have gone wrong with the classification or implementation.</Paragraph>
    <Paragraph position="8"> There are also many undisciplined locations of relational pointers (e.g. &amp;quot;@&amp;quot; and &amp;quot;~&amp;quot;, respectively 7 and 451 times in noun) in DAT files and some other problems.</Paragraph>
  </Section>
class="xml-element"></Paper>