Ranking learning-to-rank methods

D. Hiemstra, N. Tax, S. Bockting

    Onderzoeksoutput: Hoofdstuk in Boek/Rapport/CongresprocedureConferentiebijdrageAcademicpeer review

    133 Downloads (Pure)

    Samenvatting

    We present a cross-benchmark comparison of learning-to-rank methods using two evaluation measures: the Normalized Winning Number and the Ideal Winning Number. Evaluation results of 87 learning-to-rank methods on 20 datasets show that ListNet, SmoothRank, FenchelRank, FSMRank, LRUF and LARF are Pareto optimal learning-to-rank methods, listed in increasing order of Normalized Winning Number and decreasing order of Ideal Winning Number.

    Originele taal-2Engels
    TitelProceedings of the 1st International Workshop on LEARning Next gEneration Rankers
    RedacteurenN. Ferro, C. Lucchese, M. Maistro, R. Perego
    Plaats van productieAachen
    Pagina's3-3
    Aantal pagina's1
    StatusGepubliceerd - 27 nov 2017
    Evenement1st International Workshop on LEARning Next gEneration Rankers (LEARNER 2017), October 1, 2017, Amsterdam, Netherlands - Amsterdam, Nederland
    Duur: 1 okt 20171 okt 2017
    http://learner2017.dei.unipd.it/

    Publicatie series

    NaamCEUR Workshop Proceedings
    Volume2007
    ISSN van geprinte versie1613-0073

    Workshop

    Workshop1st International Workshop on LEARning Next gEneration Rankers (LEARNER 2017), October 1, 2017, Amsterdam, Netherlands
    Verkorte titelLEARNER 2017
    Land/RegioNederland
    StadAmsterdam
    Periode1/10/171/10/17
    Internet adres

    Vingerafdruk

    Duik in de onderzoeksthema's van 'Ranking learning-to-rank methods'. Samen vormen ze een unieke vingerafdruk.

    Citeer dit