<?xml version="1.0" standalone="yes"?>
<Paper uid="I05-2018">
  <Title>Detecting the Countability of English Compound Nouns Using Web-based Models</Title>
  <Section position="6" start_page="106" end_page="106" type="concl">
    <SectionTitle>
5 Conclusion
</SectionTitle>
    <Paragraph position="0"> From the results, we show that simple unsupervised web-based models can achieve the promising results on the test data. For we roughly adjusted the threshold with stepsize of 1, better performance is expected with stepsize of such as 0.1.</Paragraph>
    <Paragraph position="1"> It is unreasonable to compare the detecting results of individual and compound nouns with each other since using web-based models, compound nouns made up of two or more words are more likely to be affected by data sparseness, while individual nouns are prone to produce more noise data because of their high occurrence frequencies.</Paragraph>
    <Paragraph position="2"> Anyway using WWW is an exciting direction for NLP, how to eliminate noise data is the key to improve web-based methods. Our next step is aiming at evaluating the internet resource, distinguishing the useful and noise data.</Paragraph>
  </Section>
class="xml-element"></Paper>