<?xml version="1.0" standalone="yes"?>
<Paper uid="C94-2180">
  <Title>CUSTOMIZING AND EVALUATING A MULTILINGUAL DISCOURSE MODULE</Title>
  <Section position="6" start_page="1110" end_page="1110" type="concl">
    <SectionTitle>
5 CURRENT AND FUTURE WORK
</SectionTitle>
    <Paragraph position="0"> In this paper, we have described our multilingual discourse module mid ils customized discourse KB's, and reported the blackbox ewduation results when it was used in a data extraction task in the joint venture domain. Currently we arc working on the following two research areas in order to improve anaphora resolution.</Paragraph>
    <Paragraph position="1"> First, we are experimenting with ways to automate Iraining of anaphora resolulion by applying machine learning so that the discourse module can be castomized automaritally to a p~ticular hmguage, domain or application without extensive manual knowledge engineering. Ill order to obtain feedback liar training, we must be able to automate glassbox evaluation of discourse processing itself. For this, we have built Iwo tools: a discourse tagging tool and a discourse evaluation tool. The former has been used to tag texts with discourse relations, while tile latter lakes discourseqagged corpora as a key and the system output as results to be ev~duated.</Paragraph>
    <Paragraph position="2"> 5. According to theu approach, for a sentence &amp;quot;It was going to collect it,&amp;quot; &amp;quot;governnmnt&amp;quot; is a preferred antedeg cedent of the first &amp;quot;it,&amp;quot; while &amp;quot;money&amp;quot; is of the second, using such statistics.</Paragraph>
    <Paragraph position="3">  Second, we are expanding the rm~ge of anaphoric phenomena which our discourse module can hmldle. They include overt pronouns in English mM Spanish, and zero pronouns in Japanese and Spmfish.</Paragraph>
  </Section>
class="xml-element"></Paper>