File Information

File: 05-lr/acl_arc_1_sum/cleansed_text/xml_by_section/intro/98/w98-1120_intro.xml

Size: 1,599 bytes

Last Modified: 2025-10-06 14:06:47

<?xml version="1.0" standalone="yes"?>
<Paper uid="W98-1120">
  <Title>A Decision Tree Method for Finding and Classifying Names in Japanese Texts</Title>
  <Section position="4" start_page="171" end_page="171" type="intro">
    <SectionTitle>
2 Task
</SectionTitle>
    <Paragraph position="0"> The task is to find and classify several types of named entity items in texts, shown in Table 1.</Paragraph>
    <Paragraph position="1"> We use the task definition provided in the MET2 guidelines (Multilingual Entity Task; the formal definition will be published in May 1998).</Paragraph>
    <Paragraph position="2"> &amp;quot;Executive position&amp;quot; is a new category which is used in the portability experiment only and is not part of the MET definition.</Paragraph>
    <Paragraph position="3"> There are some idiosyncratic definitions. For  example, a sub-organization expression like &amp;quot;Executive Staff&amp;quot; should be identified only when it follows an organization in proper expression. So, in the expression &amp;quot;Defense Ministry's Executive Staff&amp;quot;, &amp;quot;Executive Staff&amp;quot; should be identified; however, it should not be identified if it appears alone in a sentence. Also, a country expression at the head of an organization should be identified if it is expressed by one Chinese character, but it should not when it is expressed by Katakana characters. Although we find some idiosyncratic definitions in the guidelines, we will use them, because there are such difficulties in nature and we can't easily find another reasonable definition.</Paragraph>
  </Section>
class="xml-element"></Paper>
Download Original XML