<?xml version="1.0" standalone="yes"?>
<Paper uid="P97-1011">
  <Title>Learning Features that Predict Cue Usage</Title>
  <Section position="3" start_page="80" end_page="80" type="relat">
    <SectionTitle>
2 Related Work
</SectionTitle>
    <Paragraph position="0"> McKeown and Elhadad (1991; 1990) studied severai connectives (e.g., but, since, because), and include many insightful hypotheses about cue selection; their observation that the distinction between but and C/lthoug/~ depends on the point of the move is related to the notion of core discussed below. However, they do not address the problem of cue occurrence.</Paragraph>
    <Paragraph position="1"> Other researchers (R6sner and Stede, 1902; Scott and de Souza, 1990) are concerned with generating text from &amp;quot;RST trees&amp;quot;, hierarchical structures where leaf nodes contain content and internal nodes indicate the rt~etorical relations, as defined in Rhetorical Structure Theory (RST) (Mann and Thompson, 1988), that exist between subtrees. They proposed heuristics for including and choosing cues based on the rhetorical relation between spans of text, the order of the relata, and the complexity of the related text spans. However, (Scott and de Souza, 1990) was based on a small number of constructed exampies, and (R6sner and Stede, 1992) focused on a small number of RST relations.</Paragraph>
    <Paragraph position="2"> (Litman, 1996) and (Siegel and McKeown, 1994) have applied machine learning to disambiguate between the discourse and sentcntial usages of cues; however, they do not consider the issues of occurrence and placement, and approach the problem from the point of view of interpretation. We closely follow the approach in (Litman, 1996) in two ways. First, we use C4.5. Second, we experiment first with each feature individually, and then with &amp;quot;interesting&amp;quot; sub-sets of features.</Paragraph>
  </Section>
class="xml-element"></Paper>