<?xml version="1.0" standalone="yes"?>
<Paper uid="P04-1043">
  <Title>A Study on Convolution Kernels for Shallow Semantic Parsing</Title>
  <Section position="6" start_page="0" end_page="0" type="concl">
    <SectionTitle>
5 Conclusions
</SectionTitle>
    <Paragraph position="0"> In this paper, we have experimented with SVMs using the two novel convolution kernels PAF and SCF which are designed for the semantic structures derived from PropBank and FrameNet corpora. Moreover, we have combined them with the polynomial kernel of standard features. The results have shown that: First, SVMs using the above kernels are appealing for semantically parsing both corpora.</Paragraph>
    <Paragraph position="1"> Second, PAF and SCF can be used to improve automatic classi cation of PropBank arguments as they provide clues about the predicate argument structure of the target verb. For example, SCF improves (a) the classi cation state-of-the-art (i.e. the polynomial kernel) of about 3 percent points and (b) the best literature result of about 5 percent points.</Paragraph>
    <Paragraph position="2"> Third, additional work is needed to design kernels suitable to learn the deep semantic contained in FrameNet as it seems not sensible to both PAF and SCF information.</Paragraph>
    <Paragraph position="3"> Finally, an analysis of SVMs using polynomial kernels over standard features has explained why they largely outperform linear classi ers based-on standard features.</Paragraph>
    <Paragraph position="4"> In the future we plan to design other structures and combine them with SCF, PAF and standard features. In this vision the learning will be carried out on a set of structural features instead of a set of at features. Other studies may relate to the use of SCF to generate verb clusters.</Paragraph>
  </Section>
class="xml-element"></Paper>