Show simple item record

dc.contributor.authorQuattoni, Ariadna Julieta
dc.contributor.authorCarreras Pérez, Xavier
dc.contributor.authorTorralba, Antonio
dc.contributor.otherUniversitat Politècnica de Catalunya. Departament de Llenguatges i Sistemes Informàtics
dc.date.accessioned2012-03-22T13:16:29Z
dc.date.available2012-03-22T13:16:29Z
dc.date.created2012
dc.date.issued2012
dc.identifier.citationQuattoni, A.J.; Carreras, X.; Torralba, A. A latent variable ranking model for content-based retrieval. A: European Conference on Information Retrieval. "Advances in Information Retrieval - 34rd European Conference on IR Research, ECIR 2012". Barcelona: Springer, 2012, p. 1-12.
dc.identifier.urihttp://hdl.handle.net/2117/15649
dc.description.abstractSince their introduction, ranking SVM models have become a powerful tool for training content-based retrieval systems. All we need for training a model are retrieval examples in the form of triplet constraints, i.e. examples specifying that relative to some query, a database item a should be ranked higher than database item b. These types of constraints could be obtained from feedback of users of the retrieval system. Most previous ranking models learn either a global combination of elementary similarity functions or a combination defined with respect to a single database item. Instead, we propose a “coarse to fine” ranking model where given a query we first compute a distribution over “coarse” classes and then use the linear combination that has been optimized for queries of that class. These coarse classes are hidden and need to be induced by the training algorithm. We propose a latent variable ranking model that induces both the latent classes and the weights of the linear combination for each class from ranking triplets. Our experiments over two large image datasets and a text retrieval dataset show the advantages of our model over learning a global combination as well as a combination for each test point (i.e. transductive setting). Furthermore, compared to the transductive approach our model has a clear computational advantages since it does not need to be retrained for each test query.
dc.format.extent12 p.
dc.language.isoeng
dc.publisherSpringer
dc.subjectÀrees temàtiques de la UPC::Informàtica::Sistemes d'informació
dc.subject.lcshRanking SVM
dc.subject.lcshInformation retrieval
dc.titleA latent variable ranking model for content-based retrieval
dc.typeConference report
dc.subject.lemacInformació -- Sistemes d'emmagatzematge i recuperació -- Models matemàtics
dc.contributor.groupUniversitat Politècnica de Catalunya. GPLN - Grup de Processament del Llenguatge Natural
dc.contributor.groupUniversitat Politècnica de Catalunya. LARCA - Laboratori d'Algorísmia Relacional, Complexitat i Aprenentatge
dc.description.peerreviewedPeer Reviewed
dc.relation.publisherversionhttp://cataleg.upc.edu/record=b1277409~S1*cat
dc.rights.accessRestricted access - publisher's policy
local.identifier.drac9411653
dc.description.versionPostprint (published version)
local.citation.authorQuattoni, A.J.; Carreras, X.; Torralba, A.
local.citation.contributorEuropean Conference on Information Retrieval
local.citation.pubplaceBarcelona
local.citation.publicationNameAdvances in Information Retrieval - 34rd European Conference on IR Research, ECIR 2012
local.citation.startingPage1
local.citation.endingPage12


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record

All rights reserved. This work is protected by the corresponding intellectual and industrial property rights. Without prejudice to any existing legal exemptions, reproduction, distribution, public communication or transformation of this work are prohibited without permission of the copyright holder