<?xml version="1.0" standalone="yes"?>
<Paper uid="W97-0317">
  <Title>Attaching Multiple Prepositional Phrases: Generalized Backed-off Estimation</Title>
  <Section position="9" start_page="153" end_page="153" type="evalu">
    <SectionTitle>
5.1 Results
</SectionTitle>
    <Paragraph position="0"> To evaluate the performance of our algorithm, we nmst first determine what the expected baseline, or lower-bound on, performance would be. Given the variation in the number of possible configurations across the three cases, the performance expected due to chance would be 50% for 1 PP, 20% for 2 PPs, and 7% for 3 PPs. A better baseline is the performance that would be expected by simply adopting the most likely configuration, without regard to lexical items. This is shown in the table below, with the most frequent configuration shown in parentheses.</Paragraph>
    <Paragraph position="2"> Table 2 presents the performance of the competitive backed-off estimation algorithm on the test data. As can be seen, the performance for PP1 replicates the findings of Collins and Brooks, who achieved 84.5% (using 4 lexical items, compared to our three). For PP2 perfordlance is again high, recalling that the algorithm is discriminating five possible attachment configurations, and the baseline expectation was only 29.8%. Similarly for PP3, our performance of 43.6% accuracy (discriminating fourteen configurations) far out strips the baseline of 18.5%.</Paragraph>
  </Section>
class="xml-element"></Paper>