<?xml version="1.0" standalone="yes"?>
<Paper uid="E95-1030">
  <Title>A fast partial parse of natural language sentences using a connectionist method</Title>
  <Section position="4" start_page="0" end_page="0" type="intro">
    <SectionTitle>
2 The corpus of sentences from
</SectionTitle>
    <Paragraph position="0"> technical manuals This work has principally been developed on text of technical manuals from Perkins Engines Ltd., which have been translated by a semi-automatic process (Pyre, 1993). Now, a partial parse can support such a process. For instance, frequently occurring modal verbs such as &amp;quot;must&amp;quot; are not dis- null tinguished by number in English, but they are in many other languages. It is necessary to locate the subject, then identify the head and determine its number in order to translate the main verb correctly in sentences like (1) below.</Paragraph>
    <Paragraph position="1"> If a cooler is fitted to the geaxl~ox, \[ the pipe \[ connections \] of the cooler \] must be regulaxly checked for corrosion. (1) This parser has been trained to find the syntactic subject head that agrees in number with the main verb. The manuals are written using the PACE (Perkins Approved Clear English) guidelines, with the aim of producing clear, unambiguous texts.</Paragraph>
    <Paragraph position="2"> AlI declarative sentences have been extracted for processing: about half were imperatives. This level of classification can be done automatically in future. Table 1 and Figure 1 show some of the characteristics of the corpus.</Paragraph>
  </Section>
class="xml-element"></Paper>