<?xml version="1.0" standalone="yes"?>
<Paper uid="A00-2015">
  <Title>Analyzing Dependencies of Japanese Subordinate Clauses based on Statistics of Scope Embedding Preference</Title>
  <Section position="4" start_page="235" end_page="235" type="relat">
    <SectionTitle>
4 Experiments and Evaluation
</SectionTitle>
    <Paragraph position="0"> We divided the 210,000 sentences of the whole EDR bracketed Japanese corpus into 95% training sentences and 5~0 test sentences. Then, we extracted 162,443 pairs of subordinate clauses from the 199,500 training sentences, and learned a decision list for dependency preference of subordinate clauses from those pairs.</Paragraph>
    <Paragraph position="1"> The default decision in the decision list is D =&amp;quot;beyond&amp;quot;, where the marginal probability</Paragraph>
    <Paragraph position="3"> precision of deciding dependency between two subordinate clauses is 53.78 %. We limit the frequency of each evidence-decision pair to be more than 9. The total number of obtained evidence-decision pairs is 7,812. We evaluate the learned decision list through several experiments. 12 First, we apply the learned decision list to deciding dependency between two subordinate clauses of the 5% test sentences. We change the threshold of the probability P(D I E) 13 in 12Details of the experimental evaluation will be presented in Utsuro (2000).</Paragraph>
    <Paragraph position="4"> I~P( D I E) can be used equivalently to the likelihood  the decision list and plot the trade-off between coverage and precision. 14 As shown in the plot of &amp;quot;Our Model&amp;quot; in Figure 5, the precision varies from 78% to 100% according to the changes of the threshold of the probability P(D I E).</Paragraph>
    <Paragraph position="5"> Next, we compare our model with the other two models: (a) the model learned by applying the decision tree learning method of Haruno et al. (1998) to our task of deciding dependency between two subordinate clauses, and (b) a decision list whose decisions are the following two cases, i.e., the case where dependency relation holds between the given two vp chunks or clauses, and the case where dependency relation does not hold. The model (b) corresponds to a model in which standard approaches to statistical dependency analysis (Collins, 1996; Fujio and Matsumoto, 1998; Haruno et al., 1998) are applied to our task of deciding dependency between two subordinate clauses. Their results are also in Figures 5 and 6. Figure 5 shows that &amp;quot;Our Model&amp;quot; outperforms the other two models in coverage. Figure 6 shows that our model outperforms both of the models (a) and (b) in coverage and precision.</Paragraph>
    <Paragraph position="6"> Finally, we examine whether the estimated dependencies of subordinate clauses improve the precision of Fujio and Matsumoto (1998)'s statistical dependency analyzer. 15 Depending on the threshold of P(D \[ E), we achieve 0.8,,~1.8% improvement in chunk level precision, and 1.6~-4.7% improvement in sentence level, is</Paragraph>
  </Section>
class="xml-element"></Paper>