<?xml version="1.0" standalone="yes"?>
<Paper uid="W06-1006">
  <Title>Violeta.Seretan@latl.unige.ch</Title>
  <Section position="7" start_page="44" end_page="46" type="concl">
    <SectionTitle>
7 Conclusion
</SectionTitle>
    <Paragraph position="0"> The paper pointed out several issues that occur in transferinga hybridcollocationextraction  methodology(thatcombineslinguisticwithstatisticinformation)toa newlanguage. Besides the questionable availability of language-specifictext analysistools for the new language,a numberof issuesthat are relevant to extraction proper were addressed: the changes in the distribution of (syntactic)word pairs, and the need to find, for each language, the most  appropriateassociationmeasureto applyforeach syntactictype (given that AMs are sensitive to distributions and syntactic types); the lack of a priori defined syntactictypes for a language; and, finally, the portabilityof some widelyused techniques(such as the window method) from English to other languages exhibiting a higher wordorderfreedom.</Paragraph>
    <Paragraph position="1"> It is again in the multilingualismperspective that the inescapableneed for preprocessingthe  textemerged(cf.differentresearcherscitedinsection 3): highlyinflectedlanguagesneed lemmatizers, free-word order languagesneed structural informationin order to guaranteeacceptableresults. As languagetoolsbecomenowadaysmore  andmoreavailable,weexpectthecollocationextraction(and terminologyacquisitionin general) to be exclusively performedin the future by relyingon linguisticanalysis. We thereforebelieve thatmultilingualismis a trueconcernforcollocationextraction. null The paper reviewed the extractionwork in a language-orientedfashion, while mentioningthe typeof linguisticpreprocessingperformedwhenever it was the case, as well as the languagespecificissues identifiedby the authors. It then proposeda strategy for implementinga multilingual extractionprocedurethat takes into account thelanguage-specificissuesidentified.</Paragraph>
    <Paragraph position="2"> An extraction system for four different languages,basedonfullparsing,wasthendescribed. null Finally, an experimentwas carriedout as a case study, whichpointedoutseveralfactorsthatmight determinea particularAMto performdifferently acrosslanguages.Theexperimentsuggestedthat log-likelihoodratios test might highlightcertain verb-objectcollocationseasierin Frenchthan in Spanish,ItalianandEnglish(in termsof salience inthesignificancelist).</Paragraph>
    <Paragraph position="3"> Futurework needsto extendthe typeof crosslinguisticanalysisinitiatedhere, in orderto providemoreinsightson the differencesexpectedat extractionbetweenonelanguageandanotherand ontheresponsiblefactors,and,accordingly, todefinesstrategiestodealwiththem. null</Paragraph>
  </Section>
class="xml-element"></Paper>