<?xml version="1.0" standalone="yes"?>
<Paper uid="J93-2006">
  <Title>BBN Systems and Technologies</Title>
  <Section position="5" start_page="378" end_page="380" type="metho">
    <SectionTitle>
DENTIAL SUMMIT WAS TO TAKE PLACE.
</SectionTitle>
    <Paragraph position="0"> Certain sequences of fragments appear frequently, as illustrated in Tables 1 and 2.</Paragraph>
    <Paragraph position="1"> One frequently occurring pair is an S followed by a PP (prepositional phrase). Since there is more than one way the parser could attach the PP, and syntactic grounds alone for attaching the PP would yield poor performance, semantic preferences applied by a post-process that combines fragments are called for.</Paragraph>
    <Paragraph position="2"> In our approach, we propose using local syntactic and semantic information rather than assuming a global syntactic and semantic form will be found. The first step is to compute a semantic interpretation for each fragment found without assuming that the meaning of each word is known. For instance, as described above, the semantic class for any noun phrase can be computed provided the head noun has semantics in the domain.</Paragraph>
    <Paragraph position="3"> Based on the data above, a reasonable approach is an algorithm that moves left-to-right through the set of fragments produced by FPP, deciding to attach fragments (or not) based on semantic criteria. To avoid requiring a complete, global analysis, a window two constituents wide is used to find patterns of possible relations among phrases. For example, an S followed by a PP invokes an action of finding all points along the &amp;quot;right edge&amp;quot; of the S tree where a PP could attach, applying the fragment combining patterns at each such spot, and ranking the alternatives.</Paragraph>
    <Paragraph position="4"> As is evident in Table 2, FPP frequently does not attach punctuation. This is to be expected, since punctuation is used in many ways, and there is no deterministic basis for attaching the constituent following the punctuation to the constituent preceding it.  Therefore, if the pair being examined by the combining algorithms ends in punctuation, the algorithm looks at the constituent following it, trying to combine it with the constituent to the left of the punctuation.</Paragraph>
    <Paragraph position="5"> A similar case is when the pair ends in a conjunction. Here the algorithm tries to combine the constituent to the right of the conjunction with that on the left of the conjunction.</Paragraph>
  </Section>
class="xml-element"></Paper>