<?xml version="1.0" standalone="yes"?>
<Paper uid="C02-2013">
  <Title>SOAT: A Semi-Automatic Domain Ontology Acquisition Tool from Chinese Corpus</Title>
  <Section position="7" start_page="0" end_page="0" type="concl">
    <SectionTitle>
5. Conclusion
</SectionTitle>
    <Paragraph position="0"> We present a semi-automatic process of domain ontology acquisition from domain corpus. The ontology schema we used is general enough for different applications and specific enough for the task of understanding the Chinese natural language. The main objective of the research is to extract useful relationships from domain articles to construct domain ontology prototypes in a semi-automatic fashion. The SOAT extraction rules we developed can identify keywords with strong semantic links, especially those compound words in the domain.</Paragraph>
    <Paragraph position="1"> We have discussed how to extract related NN pairs in Section 3 for SOAT. However, the extraction rules for NN pairs do not apply for NV pairs. In the future we shall follow the approach in (Tsai et al. 2002) to extract the relationships between nouns and its related verbs.</Paragraph>
    <Paragraph position="2"> The main restriction of SOAT is that the quality of the corpus must be very high, namely, the sentences are accurate and abundant enough to include most of the important relationships to be extracted.</Paragraph>
  </Section>
class="xml-element"></Paper>