<?xml version="1.0" standalone="yes"?>
<Paper uid="C82-1032">
  <Title>ANALYSIS AND PROCESSING OF COMPACT TEXT</Title>
  <Section position="2" start_page="0" end_page="0" type="intro">
    <SectionTitle>
INTRODUCTION
</SectionTitle>
    <Paragraph position="0"> Linguistic mechanisms of compacting are common in situations where specialists record facts to be used by others in the same field. Compact text is found in notes and records within institutions (i), and in network messages among collaborators.</Paragraph>
    <Paragraph position="1"> It is found in a different form in everyday dialogue (2), and in headings and  headlines (3).</Paragraph>
    <Paragraph position="2"> (i) Positive for heart disease and diabetes.</Paragraph>
    <Paragraph position="3"> (2) Q: How are you? Ri Fine.</Paragraph>
    <Paragraph position="4"> (3) REAGAN NAMES WOMAN JUSTICE  Similar shortened forms also occur in published research protocols and non-numeric tables (as in building codes, geological surveys). The NYU Linguistic String Project has developed a computer program to analyze and process the compact text of specialized technical areas, using a general parsing program and English grammar augmented by procedures specific to the given subject area \[1,2\]. In particular, the system has been tailored for the computer analysis of free-text medical records \[3\]. The note-taking style of medical records \[4\] uses a remarkable amount of reduced English sentence forms. For instance, in the documents reported on here, 49% of occurrences were incomplete sentences. Similar types of reductions have been described in other technical sublanguages \[5\]. In this paper, we report the results of using the LSP processor to obtain a precise description of the syntactic and semantic properties of a body of compact text.</Paragraph>
  </Section>
class="xml-element"></Paper>