<?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-03820017</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="2026-05-19T15:25:40+02:00"/>
      </publicationStmt>
      <sourceDesc>
        <p part="N">HAL API Platform</p>
      </sourceDesc>
    </fileDesc>
  </teiHeader>
  <text>
    <body>
      <listBibl>
        <biblFull>
          <titleStmt>
            <title xml:lang="en">Predictive K-means with local models</title>
            <author role="aut">
              <persName>
                <forename type="first">Vincent</forename>
                <surname>Lemaire</surname>
              </persName>
              <idno type="halauthorid">368-0</idno>
              <affiliation ref="#struct-365900"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Oumaima</forename>
                <forename type="middle">Alaoui</forename>
                <surname>Ismaili</surname>
              </persName>
              <idno type="halauthorid">2615005-0</idno>
              <affiliation ref="#struct-365900"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Antoine</forename>
                <surname>Cornuéjols</surname>
              </persName>
              <email type="md5">0a066c4d900ca96d98445ffa328a8472</email>
              <email type="domain">agroparistech.fr</email>
              <idno type="idhal" notation="string">antoine-cornuejols</idno>
              <idno type="idhal" notation="numeric">182386</idno>
              <idno type="halauthorid" notation="string">5104-182386</idno>
              <idno type="ORCID">https://orcid.org/0000-0002-2979-3521</idno>
              <idno type="IDREF">https://www.idref.fr/067132669</idno>
              <affiliation ref="#struct-1002311"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Dominique</forename>
                <surname>Gay</surname>
              </persName>
              <email type="md5">0df2068b90bdf4e3097ff2cf0ec504e9</email>
              <email type="domain">univ-reunion.fr</email>
              <idno type="idhal" notation="string">dominique-gay</idno>
              <idno type="idhal" notation="numeric">177023</idno>
              <idno type="halauthorid" notation="string">22542-177023</idno>
              <idno type="ORCID">https://orcid.org/0000-0002-0671-4616</idno>
              <idno type="IDREF">https://www.idref.fr/171034147</idno>
              <affiliation ref="#struct-54305"/>
            </author>
            <editor role="depositor">
              <persName>
                <forename>Antoine</forename>
                <surname>Cornuéjols</surname>
              </persName>
              <email type="md5">0a066c4d900ca96d98445ffa328a8472</email>
              <email type="domain">agroparistech.fr</email>
            </editor>
          </titleStmt>
          <editionStmt>
            <edition n="v1" type="current">
              <date type="whenSubmitted">2022-10-18 16:55:19</date>
              <date type="whenModified">2024-12-16 12:02:08</date>
              <date type="whenReleased">2022-10-28 11:06:01</date>
              <date type="whenProduced">2020-07-19</date>
              <date type="whenEndEmbargoed">2022-10-18</date>
              <ref type="file" target="https://hal.science/hal-03820017v1/document">
                <date notBefore="2022-10-18"/>
              </ref>
              <ref type="file" subtype="author" n="1" target="https://hal.science/hal-03820017v1/file/2012.09630.pdf" id="file-3820017-3352939">
                <date notBefore="2022-10-18"/>
              </ref>
            </edition>
            <respStmt>
              <resp>contributor</resp>
              <name key="103481">
                <persName>
                  <forename>Antoine</forename>
                  <surname>Cornuéjols</surname>
                </persName>
                <email type="md5">0a066c4d900ca96d98445ffa328a8472</email>
                <email type="domain">agroparistech.fr</email>
              </name>
            </respStmt>
          </editionStmt>
          <publicationStmt>
            <distributor>CCSD</distributor>
            <idno type="halId">hal-03820017</idno>
            <idno type="halUri">https://hal.science/hal-03820017</idno>
            <idno type="halBibtex">lemaire:hal-03820017</idno>
            <idno type="halRefHtml">&lt;i&gt;International Joint Conference on Neural Networks (IJCNN) 2020&lt;/i&gt;, Jul 2020, Glasgow, United Kingdom</idno>
            <idno type="halRef">International Joint Conference on Neural Networks (IJCNN) 2020, Jul 2020, Glasgow, United Kingdom</idno>
            <availability status="restricted">
              <licence target="https://about.hal.science/hal-authorisation-v1/">HAL Authorization<ref corresp="#file-3820017-3352939"/></licence>
            </availability>
          </publicationStmt>
          <seriesStmt>
            <idno type="stamp" n="AGROPARISTECH">AgroParisTech</idno>
            <idno type="stamp" n="AFRIQ">HAL-Francophonie, Afrique et océan indien</idno>
            <idno type="stamp" n="UNIV-REUNION">Université de la Réunion</idno>
            <idno type="stamp" n="LIM">Laboratoire d'Informatique et de Mathématiques</idno>
            <idno type="stamp" n="MIA-PARIS">MIA-Paris</idno>
            <idno type="stamp" n="UNIV-PARIS-SACLAY">Université Paris-Saclay</idno>
            <idno type="stamp" n="AGREENIUM">Archive ouverte en agrobiosciences</idno>
            <idno type="stamp" n="AGROPARISTECH-MMIP" corresp="AGROPARISTECH-ORG">AgroParisTech-MMIP</idno>
            <idno type="stamp" n="AGROPARISTECH-ORG">AgroParisTech-ORG</idno>
            <idno type="stamp" n="PAPANGUE">Papangue</idno>
            <idno type="stamp" n="FST-REUNION">UFR Sciences et technologies - Université de La Réunion</idno>
            <idno type="stamp" n="INRAE">Institut National de Recherche en Agriculture, Alimentation et Environnement</idno>
            <idno type="stamp" n="UNIVERSITE-PARIS-SACLAY" corresp="UNIV-PARIS-SACLAY">Université Paris-Saclay</idno>
            <idno type="stamp" n="GS-MATHEMATIQUES">Graduate School Mathématiques</idno>
            <idno type="stamp" n="GS-COMPUTER-SCIENCE">Graduate School Computer Science</idno>
            <idno type="stamp" n="MATHNUM">Département MathNum</idno>
            <idno type="stamp" n="RESEAU-EAU">Réseau "Systèmes Agricoles et Eau"</idno>
            <idno type="stamp" n="TEST-MATHNUM">Test MathNum</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="en">Predictive K-means with local models</title>
                <author role="aut">
                  <persName>
                    <forename type="first">Vincent</forename>
                    <surname>Lemaire</surname>
                  </persName>
                  <idno type="halauthorid">368-0</idno>
                  <affiliation ref="#struct-365900"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Oumaima</forename>
                    <forename type="middle">Alaoui</forename>
                    <surname>Ismaili</surname>
                  </persName>
                  <idno type="halauthorid">2615005-0</idno>
                  <affiliation ref="#struct-365900"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Antoine</forename>
                    <surname>Cornuéjols</surname>
                  </persName>
                  <email type="md5">0a066c4d900ca96d98445ffa328a8472</email>
                  <email type="domain">agroparistech.fr</email>
                  <idno type="idhal" notation="string">antoine-cornuejols</idno>
                  <idno type="idhal" notation="numeric">182386</idno>
                  <idno type="halauthorid" notation="string">5104-182386</idno>
                  <idno type="ORCID">https://orcid.org/0000-0002-2979-3521</idno>
                  <idno type="IDREF">https://www.idref.fr/067132669</idno>
                  <affiliation ref="#struct-1002311"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Dominique</forename>
                    <surname>Gay</surname>
                  </persName>
                  <email type="md5">0df2068b90bdf4e3097ff2cf0ec504e9</email>
                  <email type="domain">univ-reunion.fr</email>
                  <idno type="idhal" notation="string">dominique-gay</idno>
                  <idno type="idhal" notation="numeric">177023</idno>
                  <idno type="halauthorid" notation="string">22542-177023</idno>
                  <idno type="ORCID">https://orcid.org/0000-0002-0671-4616</idno>
                  <idno type="IDREF">https://www.idref.fr/171034147</idno>
                  <affiliation ref="#struct-54305"/>
                </author>
              </analytic>
              <monogr>
                <meeting>
                  <title>International Joint Conference on Neural Networks (IJCNN) 2020</title>
                  <date type="start">2020-07-19</date>
                  <settlement>Glasgow</settlement>
                  <country key="GB">United Kingdom</country>
                </meeting>
                <imprint>
                  <date type="datePub">2020</date>
                </imprint>
              </monogr>
              <relatedItem target="https://hal.univ-reunion.fr/hal-04500406v1" type="IsVersionOf" subtype="http://purl.org/coar/resource_type/c_18cf"/>
            </biblStruct>
          </sourceDesc>
          <profileDesc>
            <langUsage>
              <language ident="en">English</language>
            </langUsage>
            <textClass>
              <classCode scheme="halDomain" n="info.info-lg">Computer Science [cs]/Machine Learning [cs.LG]</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>Supervised classification can be effective for prediction but sometimes weak on interpretability or explainability (XAI). Clustering, on the other hand, tends to isolate categories or profiles that can be meaningful but there is no guarantee that they are useful for labels prediction. Predictive clustering seeks to obtain the best of the two worlds. Starting from labeled data, it looks for clusters that are as pure as possible with regards to the class labels. One technique consists in tweaking a clustering algorithm so that data points sharing the same label tend to aggregate together. With distance-based algorithms, such as k-means, a solution is to modify the distance used by the algorithm so that it incorporates information about the labels of the data points. In this paper, we propose another method which relies on a change of representation guided by class densities and then carries out clustering in this new representation space. We present two new algorithms using this technique and show on a variety of data sets that they are competitive for prediction performance with pure supervised classifiers while offering interpretability of the clusters discovered.</p>
            </abstract>
            <abstract xml:lang="fr">
              <p>La classification supervisée peut être efficace pour la prédiction mais parfois insatisfaisante en matière d'interprétabilité ou d'explicabilité (XAI). Le clustering, quant à lui, tend à isoler des catégories ou des profils qui peuvent être interprétables mais il n'y a aucune garantie qu'ils soient utiles pour la prédiction d'étiquettes. Le clustering prédictif cherche à obtenir le meilleur des deux mondes. À partir de données étiquetées, il recherche des clusters qui sont aussi purs que possible en ce qui concerne les étiquettes de classe. Une technique consiste à modifier un algorithme de clustering pour que les points de données partageant la même étiquette aient tendance à s'agréger ensemble. Avec les algorithmes basés sur la distance, comme les k-means, une solution consiste à modifier la distance utilisée par l'algorithme afin qu'elle intègre des informations sur les étiquettes des points de données. Dans cet article, nous proposons une autre méthode qui s'appuie sur un changement de représentation guidé par les densités de classes, puis effectue le clustering dans ce nouvel espace de représentation. Nous présentons deux nouveaux algorithmes utilisant cette technique et montrons sur une variété de jeux de données qu'ils sont compétitifs pour la performance de prédiction avec les classifieurs supervisés purs tout en offrant une interprétabilité des clusters découverts.</p>
            </abstract>
          </profileDesc>
        </biblFull>
      </listBibl>
    </body>
    <back>
      <listOrg type="structures">
        <org type="institution" xml:id="struct-365900" status="VALID">
          <orgName>Orange Labs</orgName>
          <desc>
            <address>
              <country key="FR"/>
            </address>
          </desc>
        </org>
        <org type="laboratory" xml:id="struct-1002311" status="VALID">
          <idno type="RNSR">201119642G</idno>
          <idno type="ROR">https://ror.org/00780az30</idno>
          <orgName>Mathématiques et Informatique Appliquées</orgName>
          <orgName type="acronym">MIA Paris-Saclay</orgName>
          <date type="start">2020-01-01</date>
          <desc>
            <address>
              <addrLine>22 place de l'Agronomie CS 20040 91123 Palaiseau cedex</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">https://mia-ps.inrae.fr/</ref>
          </desc>
          <listRelation>
            <relation active="#struct-148117" type="direct"/>
            <relation active="#struct-419361" type="direct"/>
            <relation name="UMR0518" active="#struct-577435" type="direct"/>
          </listRelation>
        </org>
        <org type="laboratory" xml:id="struct-54305" status="VALID">
          <idno type="RNSR">199814054P</idno>
          <orgName>Laboratoire d'Informatique et de Mathématiques</orgName>
          <orgName type="acronym">LIM</orgName>
          <date type="start">2008-01-01</date>
          <desc>
            <address>
              <addrLine>Faculté des Sciences et Technologies - Université de la Réunion, 2, rue Joseph Wetzell 97490 Sainte-Clotilde</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">http://lim.univ-reunion.fr/</ref>
          </desc>
          <listRelation>
            <relation name="EA2525" active="#struct-300316" type="direct"/>
          </listRelation>
        </org>
        <org type="institution" xml:id="struct-148117" status="VALID">
          <idno type="IdRef">139408088</idno>
          <idno type="ROR">https://ror.org/02kbmgc12</idno>
          <orgName>AgroParisTech</orgName>
          <date type="start">2007-01-01</date>
          <desc>
            <address>
              <addrLine>22 place de l'Agronomie CS 20040 91123 Palaiseau cedex</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">https://www.agroparistech.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-577435" status="VALID">
          <idno type="ROR">https://ror.org/003vg9w96</idno>
          <orgName>Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement</orgName>
          <orgName type="acronym">INRAE</orgName>
          <date type="start">2020-01-01</date>
          <desc>
            <address>
              <country key="FR"/>
            </address>
          </desc>
        </org>
        <org type="institution" xml:id="struct-300316" status="VALID">
          <idno type="IdRef">026404451</idno>
          <idno type="ISNI">0000000121112608</idno>
          <idno type="ROR">https://ror.org/005ypkf75</idno>
          <orgName>Université de La Réunion</orgName>
          <orgName type="acronym">UR</orgName>
          <date type="start">1982-10-15</date>
          <desc>
            <address>
              <addrLine>15, avenue René Cassin - CS92003 -97744 Saint Denis Cedex 9</addrLine>
              <country key="RE"/>
            </address>
            <ref type="url">http://www.univ-reunion.fr</ref>
          </desc>
        </org>
      </listOrg>
    </back>
  </text>
</TEI>