<?xml version="1.0" standalone="yes"?>
<Paper uid="W04-2414">
  <Title>Memory-based semantic role labeling: Optimizing features, algorithm, and output</Title>
  <Section position="5" start_page="0" end_page="0" type="concl">
    <SectionTitle>
5 Conclusion
</SectionTitle>
    <Paragraph position="0"> We have described a memory-based semantic role labeler.</Paragraph>
    <Paragraph position="1"> In the development of the system we have used feature selection through bi-directional hill-climbing and parameter optimization through iterative deepening search. We have evaluated n-gram output classes, classifier stacking and output post-processing, all of which increased performance. An overview of the performance of the system on the test data can be found in Table 4.</Paragraph>
  </Section>
class="xml-element"></Paper>