<?xml version="1.0" standalone="yes"?>
<Paper uid="C96-1003">
  <Title>Clustering Words with the MDL Principle</Title>
  <Section position="7" start_page="7" end_page="7" type="concl">
    <SectionTitle>
6 Conchtding Remarks
</SectionTitle>
    <Paragraph position="0"> We have proposed a tnethod of&amp;quot; hierarchical &lt;'hisfeting of words hased on laxge corpus data. \Vo conclude wit, h the following remarl,:s, \[. ()lip ll/et ho(\[ (&gt;\[&amp;quot; chtst:('ritlg w(wds has(,d cqt th(' MI)L t',ritlciph&amp;quot; is ~h&lt;,reficalty sc.+,rtd. Our experimental t'esult.s show t.hal: il. is I+ot t.er to enq)loy MI)I, than M 1,1!; as estimation &lt;'riWrion in hierarchical word chtstering.</Paragraph>
    <Paragraph position="1">  2. \[lsing a tlwsaltrus consl.rt~cl,cd l&gt;y ottr met hod can inq&gt;rov(; pp-;fl.t, ach)nent, disaml)igtmtion results.</Paragraph>
    <Paragraph position="2"> 3. At, t.he Clm:ent, st, a, te of the art. itt st.al, istical  na.t;ttral languag(~ I)rocessing, it. is b('st. I.o use a cotnbination of an a.ut.ontat.ically const.rucl(&gt;d thesa.urus and a hand made l;hesattrus \['or disatnbigua.l.ion purpose. &amp;quot;Fhe disaulhiglmtion accttra.cy obtained this way wets 85+5(/c,.</Paragraph>
    <Paragraph position="3"> \[u the fut.ttre, hopefillly wit, h target training dat.a size, we plao_ to construct, larger thesauri as well as to test other clustering algorit.hms.</Paragraph>
  </Section>
class="xml-element"></Paper>