<?xml version="1.0" standalone="yes"?>
<Paper uid="P98-2186">
  <Title>Part of Speech Tagging Using a Network of Linear Separators</Title>
  <Section position="7" start_page="1141" end_page="1141" type="concl">
    <SectionTitle>
6 Conclusion
</SectionTitle>
    <Paragraph position="0"> A Winnow-based network of linear separators was shown to be very effective when applied to POS tagging. We described the SNOW architecture and how to use it for POS tagging and found that although the algorithm is an on-line algorithm, with the advantages this carries, its performance is comparable to the best taggers available.</Paragraph>
    <Paragraph position="1"> This work opens a variety of questions. Some are related to further studying this approach, based on multiplicative update algorithms, and using it for other natural language problems.</Paragraph>
    <Paragraph position="2"> More fundamental, we believe, are those that are concerned with the general learning paradigm the SNOW architecture proposes.</Paragraph>
    <Paragraph position="3"> A large number of different kinds of ambiguities are to be resolved simultaneously in performing any higher level natural language inference (Cardie, 1996). Naturally, these processes, acting on the same input and using the same &amp;quot;memory&amp;quot;, will interact. In SNO W, a collection of classifiers are used; all are learned from the same data, and share the same &amp;quot;memory&amp;quot;. In the study of SNOWwe embark on the study of some of the fundamental issues that are involved in putting together a large number of classifiers and investigating the interactions among them, with the hope of making progress towards using these in performing higher level inferences.</Paragraph>
  </Section>
class="xml-element"></Paper>