<?xml version="1.0" standalone="yes"?>
<Paper uid="H91-1065">
  <Title>STUDIES IN PART OF SPEECH LABELLING</Title>
  <Section position="9" start_page="335" end_page="335" type="concl">
    <SectionTitle>
7. FUTURE DIRECTIONS
</SectionTitle>
    <Paragraph position="0"> In the work reported here, we have evaluated POST in the laboratory, comparing its results against the work of people doing the same task. However, the real test of such a system is how well it functions as a component in a larger system. Can it make a parser work faster and more accurately? Can it help to extract certain kinds of phrases from unrestricted text? We are currently running these experiments by making POST a part of existing systems. It is being run as a preprocessor to Gfishman's Proteus system for the MUC-3 competition (Gnshman &amp; Sterling 1989).</Paragraph>
    <Paragraph position="1"> Preliminary results showed it sped up Proteus by a factor of two in one-best mode and by a factor of 33% with a threshold of T=2. It is also being integrated into a new message processing system at BBN. The results of these experiments will provide us with new directions and ideas both for improving POST and for other ways to integrate probabilistic models into natural language processing systems.</Paragraph>
  </Section>
class="xml-element"></Paper>