<?xml version="1.0" encoding="utf-8"?>
<TEI xmlns="http://www.tei-c.org/ns/1.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:hal="http://hal.archives-ouvertes.fr/" xmlns:gml="http://www.opengis.net/gml/3.3/" xmlns:gmlce="http://www.opengis.net/gml/3.3/ce" version="1.1" xsi:schemaLocation="http://www.tei-c.org/ns/1.0 http://api.archives-ouvertes.fr/documents/aofr-sword.xsd">
  <teiHeader>
    <fileDesc>
      <titleStmt>
        <title>HAL TEI export of hal-02432990</title>
      </titleStmt>
      <publicationStmt>
        <distributor>CCSD</distributor>
        <availability status="restricted">
          <licence target="https://creativecommons.org/publicdomain/zero/1.0/">CC0 1.0 - Universal</licence>
        </availability>
        <date when="2026-05-24T21:42:39+02:00"/>
      </publicationStmt>
      <sourceDesc>
        <p part="N">HAL API Platform</p>
      </sourceDesc>
    </fileDesc>
  </teiHeader>
  <text>
    <body>
      <listBibl>
        <biblFull>
          <titleStmt>
            <title xml:lang="fr">Représentations lexicales pour la détection non supervisée d'événements dans un flux de tweets : étude sur des corpus français et anglais</title>
            <author role="aut">
              <persName>
                <forename type="first">Béatrice</forename>
                <surname>Mazoyer</surname>
              </persName>
              <email type="md5">e265a05c3d0c93843c825b0b0cb751b5</email>
              <email type="domain">sciencespo.fr</email>
              <idno type="idhal" notation="string">beatrice-mazoyer</idno>
              <idno type="idhal" notation="numeric">178192</idno>
              <idno type="halauthorid" notation="string">46596-178192</idno>
              <idno type="ORCID">https://orcid.org/0000-0003-4074-2976</idno>
              <idno type="IDREF">https://www.idref.fr/250144468</idno>
              <affiliation ref="#struct-1042590"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Nicolas</forename>
                <surname>Hervé</surname>
              </persName>
              <email type="md5">f4413e0ae14d4b17bc0c699a0526f104</email>
              <email type="domain">ina.fr</email>
              <idno type="idhal" notation="numeric">948944</idno>
              <idno type="halauthorid" notation="string">121388-948944</idno>
              <affiliation ref="#struct-300215"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Céline</forename>
                <surname>Hudelot</surname>
              </persName>
              <email type="md5">f47c8e3cb17a2bf7d43f2c152bb86885</email>
              <email type="domain">centralesupelec.fr</email>
              <idno type="idhal" notation="string">celine-hudelot</idno>
              <idno type="idhal" notation="numeric">20877</idno>
              <idno type="halauthorid" notation="string">30850-20877</idno>
              <idno type="ORCID">https://orcid.org/0000-0003-3849-4133</idno>
              <idno type="ARXIV">https://arxiv.org/a/hudelot_c_1</idno>
              <idno type="GOOGLE SCHOLAR">https://scholar.google.fr/citations?user=gFlAh6MAAAAJ</idno>
              <idno type="IDREF">https://www.idref.fr/113366744</idno>
              <affiliation ref="#struct-1042590"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Julia</forename>
                <surname>Cage</surname>
              </persName>
              <email type="md5">e6abfe7d59e4e925d02979e4d2f1ac56</email>
              <email type="domain">sciences-po.org</email>
              <ptr type="url" target="https://juliacage.com/"/>
              <idno type="idhal" notation="string">julia-cage</idno>
              <idno type="idhal" notation="numeric">1101763</idno>
              <idno type="halauthorid" notation="string">1153773-1101763</idno>
              <idno type="RESEARCHERID">http://www.researcherid.com/rid/AAA-1522-2019</idno>
              <idno type="ORCID">https://orcid.org/0000-0001-7085-5774</idno>
              <idno type="IDREF">https://www.idref.fr/160569524</idno>
              <idno type="RESEARCHERID">http://www.researcherid.com/rid/http://www.researcherid.com/rid/AAA-1522-2019</idno>
              <affiliation ref="#struct-226874"/>
            </author>
            <editor role="depositor">
              <persName>
                <forename>Béatrice</forename>
                <surname>Mazoyer</surname>
              </persName>
              <email type="md5">e265a05c3d0c93843c825b0b0cb751b5</email>
              <email type="domain">sciencespo.fr</email>
            </editor>
          </titleStmt>
          <editionStmt>
            <edition n="v1" type="current">
              <date type="whenSubmitted">2020-01-08 17:57:12</date>
              <date type="whenModified">2024-07-16 11:56:20</date>
              <date type="whenReleased">2020-01-13 11:26:07</date>
              <date type="whenProduced">2020-01-27</date>
              <date type="whenEndEmbargoed">2020-01-08</date>
              <ref type="file" target="https://centralesupelec.hal.science/hal-02432990v1/document">
                <date notBefore="2020-01-08"/>
              </ref>
              <ref type="file" subtype="author" n="1" target="https://centralesupelec.hal.science/hal-02432990v1/file/EGC_2020.pdf" id="file-2432990-2323191">
                <date notBefore="2020-01-08"/>
              </ref>
              <ref type="externalLink" target="http://arxiv.org/pdf/2001.04139"/>
            </edition>
            <respStmt>
              <resp>contributor</resp>
              <name key="801000">
                <persName>
                  <forename>Béatrice</forename>
                  <surname>Mazoyer</surname>
                </persName>
                <email type="md5">e265a05c3d0c93843c825b0b0cb751b5</email>
                <email type="domain">sciencespo.fr</email>
              </name>
            </respStmt>
          </editionStmt>
          <publicationStmt>
            <distributor>CCSD</distributor>
            <idno type="halId">hal-02432990</idno>
            <idno type="halUri">https://centralesupelec.hal.science/hal-02432990</idno>
            <idno type="halBibtex">mazoyer:hal-02432990</idno>
            <idno type="halRefHtml">&lt;i&gt;Extraction et Gestion des connaissances, EGC 2020&lt;/i&gt;, Jan 2020, Bruxelles, Belgique</idno>
            <idno type="halRef">Extraction et Gestion des connaissances, EGC 2020, Jan 2020, Bruxelles, Belgique</idno>
            <availability status="restricted">
              <licence target="https://about.hal.science/hal-authorisation-v1/">HAL Authorization<ref corresp="#file-2432990-2323191"/></licence>
            </availability>
          </publicationStmt>
          <seriesStmt>
            <idno type="stamp" n="SCIENCESPO">Sciences Po</idno>
            <idno type="stamp" n="CNRS">CNRS - Centre national de la recherche scientifique</idno>
            <idno type="stamp" n="CENTRALESUPELEC">Ecole CentraleSupélec</idno>
            <idno type="stamp" n="MICS">Mathématiques et Informatique pour la Complexité et les Systèmes</idno>
            <idno type="stamp" n="USPC">Université Sorbonne Paris Cité</idno>
            <idno type="stamp" n="UNIV-PARIS-SACLAY">Université Paris-Saclay</idno>
            <idno type="stamp" n="TEST-HALCNRS">Collection test HAL CNRS</idno>
            <idno type="stamp" n="UNIVERSITE-PARIS-SACLAY" corresp="UNIV-PARIS-SACLAY">Université Paris-Saclay</idno>
            <idno type="stamp" n="GS-ENGINEERING">Graduate School Sciences de l'Ingénierie et des Systèmes</idno>
            <idno type="stamp" n="GS-COMPUTER-SCIENCE">Graduate School Computer Science</idno>
            <idno type="stamp" n="HUB-IA">Hub Intelligence Artificielle de CentraleSupélec</idno>
            <idno type="stamp" n="ECON-SCPO" corresp="SCIENCESPO">Les publications du Département d'économie de Sciences Po</idno>
            <idno type="stamp" n="SCPO_OA" corresp="SCIENCESPO">Sciences Po Open Access</idno>
          </seriesStmt>
          <notesStmt>
            <note type="audience" n="2">International</note>
            <note type="invited" n="0">No</note>
            <note type="popular" n="0">No</note>
            <note type="peer" n="1">Yes</note>
            <note type="proceedings" n="0">No</note>
          </notesStmt>
          <sourceDesc>
            <biblStruct>
              <analytic>
                <title xml:lang="fr">Représentations lexicales pour la détection non supervisée d'événements dans un flux de tweets : étude sur des corpus français et anglais</title>
                <author role="aut">
                  <persName>
                    <forename type="first">Béatrice</forename>
                    <surname>Mazoyer</surname>
                  </persName>
                  <email type="md5">e265a05c3d0c93843c825b0b0cb751b5</email>
                  <email type="domain">sciencespo.fr</email>
                  <idno type="idhal" notation="string">beatrice-mazoyer</idno>
                  <idno type="idhal" notation="numeric">178192</idno>
                  <idno type="halauthorid" notation="string">46596-178192</idno>
                  <idno type="ORCID">https://orcid.org/0000-0003-4074-2976</idno>
                  <idno type="IDREF">https://www.idref.fr/250144468</idno>
                  <affiliation ref="#struct-1042590"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Nicolas</forename>
                    <surname>Hervé</surname>
                  </persName>
                  <email type="md5">f4413e0ae14d4b17bc0c699a0526f104</email>
                  <email type="domain">ina.fr</email>
                  <idno type="idhal" notation="numeric">948944</idno>
                  <idno type="halauthorid" notation="string">121388-948944</idno>
                  <affiliation ref="#struct-300215"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Céline</forename>
                    <surname>Hudelot</surname>
                  </persName>
                  <email type="md5">f47c8e3cb17a2bf7d43f2c152bb86885</email>
                  <email type="domain">centralesupelec.fr</email>
                  <idno type="idhal" notation="string">celine-hudelot</idno>
                  <idno type="idhal" notation="numeric">20877</idno>
                  <idno type="halauthorid" notation="string">30850-20877</idno>
                  <idno type="ORCID">https://orcid.org/0000-0003-3849-4133</idno>
                  <idno type="ARXIV">https://arxiv.org/a/hudelot_c_1</idno>
                  <idno type="GOOGLE SCHOLAR">https://scholar.google.fr/citations?user=gFlAh6MAAAAJ</idno>
                  <idno type="IDREF">https://www.idref.fr/113366744</idno>
                  <affiliation ref="#struct-1042590"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Julia</forename>
                    <surname>Cage</surname>
                  </persName>
                  <email type="md5">e6abfe7d59e4e925d02979e4d2f1ac56</email>
                  <email type="domain">sciences-po.org</email>
                  <ptr type="url" target="https://juliacage.com/"/>
                  <idno type="idhal" notation="string">julia-cage</idno>
                  <idno type="idhal" notation="numeric">1101763</idno>
                  <idno type="halauthorid" notation="string">1153773-1101763</idno>
                  <idno type="RESEARCHERID">http://www.researcherid.com/rid/AAA-1522-2019</idno>
                  <idno type="ORCID">https://orcid.org/0000-0001-7085-5774</idno>
                  <idno type="IDREF">https://www.idref.fr/160569524</idno>
                  <idno type="RESEARCHERID">http://www.researcherid.com/rid/http://www.researcherid.com/rid/AAA-1522-2019</idno>
                  <affiliation ref="#struct-226874"/>
                </author>
              </analytic>
              <monogr>
                <meeting>
                  <title>Extraction et Gestion des connaissances, EGC 2020</title>
                  <date type="start">2020-01-27</date>
                  <settlement>Bruxelles</settlement>
                  <country key="BE">Belgium</country>
                </meeting>
                <imprint/>
              </monogr>
              <idno type="arxiv">2001.04139</idno>
            </biblStruct>
          </sourceDesc>
          <profileDesc>
            <langUsage>
              <language ident="fr">French</language>
            </langUsage>
            <textClass>
              <classCode scheme="halDomain" n="info.info-tt">Computer Science [cs]/Document and Text Processing</classCode>
              <classCode scheme="halDomain" n="info.info-ir">Computer Science [cs]/Information Retrieval [cs.IR]</classCode>
              <classCode scheme="halDomain" n="info.info-si">Computer Science [cs]/Social and Information Networks [cs.SI]</classCode>
              <classCode scheme="halTypology" n="COMM">Conference papers</classCode>
              <classCode scheme="halOldTypology" n="COMM">Conference papers</classCode>
              <classCode scheme="halTreeTypology" n="COMM">Conference papers</classCode>
            </textClass>
            <abstract xml:lang="en">
              <p>In this work, we evaluate the performance of recent text embeddings for the automatic detection of events in a stream of tweets. We model this task as a dynamic clustering problem.Our experiments are conducted on a publicly available corpus of tweets in English and on a similar dataset in French annotated by our team. We show that recent techniques based on deep neural networks (ELMo, Universal Sentence Encoder, BERT, SBERT), although promising on many applications, are not very suitable for this task. We also experiment with different types of fine-tuning to improve these results on French data. Finally, we propose a detailed analysis of the results obtained, showing the superiority of tf-idf approaches for this task.</p>
            </abstract>
            <abstract xml:lang="fr">
              <p>Dans cet article, nous nous intéressons aux approches récentes de plongements lexicaux en vue de les appliquer à la détection automatique d'événements dans un flux de tweets. Nous modélisons cette tâche comme un problème de clustering dynamique. Nos expériences sont menées sur un corpus de tweets en anglais accessible publiquement ainsi que sur un jeu de données similaire en français annoté par notre équipe. Nous montrons que les techniques récentes fondées sur des réseaux de neurones profonds (ELMo, Universal Sentence Encoder, BERT, SBERT), bien que prometteuses sur de nombreuses applications , sont peu adaptées pour cette tâche, même sur le corpus en anglais. Nous expérimentons également différents types de fine-tuning afin d'améliorer les résultats de ces modèles sur les données en français. Nous proposons enfin une analyse fine des résultats obtenus montrant la supériorité des approches traditionnelles de type tf-idf pour ce type de tâche et de corpus.</p>
            </abstract>
          </profileDesc>
        </biblFull>
      </listBibl>
    </body>
    <back>
      <listOrg type="structures">
        <org type="laboratory" xml:id="struct-1042590" status="VALID">
          <idno type="IdRef">14797299X</idno>
          <idno type="RNSR">200615286U</idno>
          <idno type="ROR">https://ror.org/03q29s414</idno>
          <orgName>Mathématiques et Informatique pour la Complexité et les Systèmes</orgName>
          <orgName type="acronym">MICS</orgName>
          <date type="start">2020-01-01</date>
          <desc>
            <address>
              <addrLine>Plateau de Moulon 3 rue Joliot Curie 91192 GIF SUR YVETTE CEDEX</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">http://mics.centralesupelec.fr/</ref>
          </desc>
          <listRelation>
            <relation active="#struct-411575" type="direct"/>
            <relation active="#struct-419361" type="direct"/>
          </listRelation>
        </org>
        <org type="regroupinstitution" xml:id="struct-300215" status="VALID">
          <idno type="ROR">https://ror.org/02zwkwz55</idno>
          <orgName>Institut National de l'Audiovisuel</orgName>
          <orgName type="acronym">INA</orgName>
          <desc>
            <address>
              <addrLine>4 avenue de l'Europe, 94366 Bry-sur-Marne Cedex</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">http://www.institut-national-audiovisuel.fr/</ref>
          </desc>
        </org>
        <org type="laboratory" xml:id="struct-226874" status="VALID">
          <idno type="RNSR">200820915D</idno>
          <orgName>Département d'économie (Sciences Po)</orgName>
          <orgName type="acronym">ECON</orgName>
          <desc>
            <address>
              <addrLine>28 rue des Saints-Pères - 75007 Paris</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">http://econ.sciences-po.fr/</ref>
          </desc>
          <listRelation>
            <relation active="#struct-301587" type="direct"/>
            <relation name="UMR8259" active="#struct-441569" type="direct"/>
          </listRelation>
        </org>
        <org type="institution" xml:id="struct-411575" status="VALID">
          <idno type="IdRef">184443237</idno>
          <idno type="ROR">https://ror.org/019tcpt25</idno>
          <orgName>CentraleSupélec</orgName>
          <desc>
            <address>
              <addrLine>3, rue Joliot Curie,Plateau de Moulon,91192 GIF-SUR-YVETTE Cedex</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">http://www.centralesupelec.fr</ref>
          </desc>
        </org>
        <org type="institution" xml:id="struct-419361" status="VALID">
          <idno type="IdRef">241345251</idno>
          <idno type="ROR">https://ror.org/03xjwb503</idno>
          <orgName>Université Paris-Saclay</orgName>
          <desc>
            <address>
              <addrLine>Bâtiment Bréguet, 3 Rue Joliot Curie 2e ét, 91190 Gif-sur-Yvette</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">https://www.universite-paris-saclay.fr/fr</ref>
          </desc>
        </org>
        <org type="institution" xml:id="struct-301587" status="VALID">
          <idno type="IdRef">027918459</idno>
          <idno type="ROR">https://ror.org/05fe7ax82</idno>
          <orgName>Sciences Po</orgName>
          <orgName type="acronym">Sciences Po</orgName>
          <desc>
            <address>
              <addrLine>27, rue Saint-Guillaume - 75337 Paris cedex 07</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">https://www.sciencespo.fr/</ref>
          </desc>
        </org>
        <org type="regroupinstitution" xml:id="struct-441569" status="VALID">
          <idno type="IdRef">02636817X</idno>
          <idno type="ISNI">0000000122597504</idno>
          <idno type="ROR">https://ror.org/02feahw73</idno>
          <orgName>Centre National de la Recherche Scientifique</orgName>
          <orgName type="acronym">CNRS</orgName>
          <date type="start">1939-10-19</date>
          <desc>
            <address>
              <country key="FR"/>
            </address>
            <ref type="url">https://www.cnrs.fr/</ref>
          </desc>
        </org>
      </listOrg>
    </back>
  </text>
</TEI>