<?xml version="1.0" standalone="yes"?>
<Paper uid="W00-0731">
  <Title>Shallow Parsing as Part-of-Speech Tagging*</Title>
  <Section position="7" start_page="146" end_page="146" type="concl">
    <SectionTitle>
6 Comments
</SectionTitle>
    <Paragraph position="0"> As was argued in the introduction, increasing the size of the context produces better results, and such performance is bounded by issues such as sparse statistics. Our experiments suggest that this was indeed true.</Paragraph>
    <Paragraph position="1"> We make no claims about the generality of our modelling. Clearly it is specific to the tagger used.</Paragraph>
    <Paragraph position="2"> In more detail, we found that: * PPs seem easy to identify.</Paragraph>
    <Paragraph position="3"> * ADJP and ADVP chunks were hard to identify correctly. We suspect that improvements here require greater syntactic information than just base-phrases.</Paragraph>
    <Paragraph position="4"> * Our performance at NPs should be improved-upon. In terms of modelling, we did not treat any chunk differently from any other chunk. We also did not treat any words differently from any other words.</Paragraph>
    <Paragraph position="5"> * The performance using just words and just POS tags were roughly equivalent. However, the performance using both sources was better than when using either source of information in isolation. The reason for this is that words and POS tags have different properties, and that together, the specificity of words can overcome the coarseness of tags, whilst the abundance of tags can deal with the sparseness of words.</Paragraph>
    <Paragraph position="6"> Our results were not wildly worse than those reported by Buchholz et al (Sabine Buchholz and Daelemans, 1999). This comparable level of performance suggests that shallow parsing (base  accuracy: 94.88% phrasal recognition) is a fairly easy task. Improvements might come from better modelling, dealing with illegal chunk sequences, allowing multiple chunks with confidence intervals, system combination etc, but we feel that such improvements will be small. Given this, we believe that base-phrasal chunking is close to being a solved problem.</Paragraph>
  </Section>
class="xml-element"></Paper>