We derive the clustering problem from first principles showing that the goal of achieving a probabilistic, or "hard", multi class clustering result is equivalent to the algebraic problem of a completely positive factorization under a doubly stochastic constraint. We show that spectral clustering, normalized cuts, kernel K-means and the various normalizations of the associated affinity matrix are particular instances and approximations of this general principle. We propose an efficient algorithm for achieving a completely positive factorization and extend the basic clustering scheme to situations where partial label information is available.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    A unifying approach to hard and probabilistic clustering


    Beteiligte:
    Zass, R. (Autor:in) / Shashua, A. (Autor:in)


    Erscheinungsdatum :

    2005-01-01


    Format / Umfang :

    400491 byte





    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    A Unifying Approach to Hard and Probabilistic Clustering

    Zass, R. / Shashua, A. / IEEE | British Library Conference Proceedings | 2005



    A unifying approach to multitarget tracking

    Emre, E. / Seo, J.H. | IEEE | 1989


    Unifying HPC

    Bales, Bryan | NTRS | 2017


    FNS, CFNS and HEIV: A Unifying Approach

    Chojnacki, W. / Brooks, M. J. / Hengel, A. et al. | British Library Online Contents | 2005