<?xml version="1.0" standalone="yes"?>
<Paper uid="W06-3309">
  <Title>Generative Content Models for Structural Analysis of Medical Abstracts</Title>
  <Section position="7" start_page="70" end_page="70" type="concl">
    <SectionTitle>
6 Conclusion
</SectionTitle>
    <Paragraph position="0"> We believe that there are two contributions as a result of our work. From the perspective of machine learning, the assignment of sequentially-occurring labels represents an underexplored problem with respect to the generative vs. discriminative debate-previous work has mostly focused on stateless classification tasks. This paper demonstrates that Hidden Markov Models are capable of capturing discourse transitions from section to section, and are at least competitive with Support Vector Machines from a purely performance point of view.</Paragraph>
    <Paragraph position="1"> The other contribution of our work is that it contributes to building advanced clinical information systems. From an application point of view, the ability to assign structure to otherwise unstructured text represents a key capability that may assist in question answering, document summarization, and other natural language processing applications.</Paragraph>
    <Paragraph position="2"> Much research in computational linguistics has focused on corpora comprised of newswire articles.</Paragraph>
    <Paragraph position="3"> We would like to point out that clinical texts provide another attractive genre in which to conduct experiments. Such texts are easy to acquire, and the availability of domain ontologies provides new opportunities for knowledge-rich approaches to shine. Although we have only experimented with lexical features in this study, the door is wide open for follow-on studies based on semantic features.</Paragraph>
  </Section>
class="xml-element"></Paper>