Multi-group Support Vector Machines with Measurement Costs: A Biobjective Approach

Emilio Carrizosa, Belén Martín-Barragán, Dolores Romero Morales

Research output: Contribution to journalJournal articleResearchpeer-review

Abstract

Support Vector Machine has shown to have good performance in many practical classification settings. In this paper we propose, for multi-group classification, a biobjective optimization model in which we consider not only the generalization ability (modeled through the margin maximization), but also costs associated with the features. This cost is not limited to an economical payment, but can also refer to risk, computational effort, space requirements, etc. We introduce a Biobjective Mixed Integer Problem, for which Pareto optimal solutions are obtained. Those Pareto optimal solutions correspond to different classification rules, among which the user would choose the one yielding the most appropriate compromise between the cost and the expected misclassification rate.
Support Vector Machine has shown to have good performance in many practical classification settings. In this paper we propose, for multi-group classification, a biobjective optimization model in which we consider not only the generalization ability (modeled through the margin maximization), but also costs associated with the features. This cost is not limited to an economical payment, but can also refer to risk, computational effort, space requirements, etc. We introduce a Biobjective Mixed Integer Problem, for which Pareto optimal solutions are obtained. Those Pareto optimal solutions correspond to different classification rules, among which the user would choose the one yielding the most appropriate compromise between the cost and the expected misclassification rate.
LanguageEnglish
JournalDiscrete Applied Mathematics
Volume156
Issue number6
Pages950–966
ISSN0166-218X
DOIs
StatePublished - 2008
Externally publishedYes

Keywords

    Cite this

    Carrizosa, Emilio ; Martín-Barragán, Belén ; Morales, Dolores Romero. / Multi-group Support Vector Machines with Measurement Costs : A Biobjective Approach. In: Discrete Applied Mathematics. 2008 ; Vol. 156, No. 6. pp. 950–966
    @article{0e9a5dee3b0848f090f69f082ca0bdf0,
    title = "Multi-group Support Vector Machines with Measurement Costs: A Biobjective Approach",
    abstract = "Support Vector Machine has shown to have good performance in many practical classification settings. In this paper we propose, for multi-group classification, a biobjective optimization model in which we consider not only the generalization ability (modeled through the margin maximization), but also costs associated with the features. This cost is not limited to an economical payment, but can also refer to risk, computational effort, space requirements, etc. We introduce a Biobjective Mixed Integer Problem, for which Pareto optimal solutions are obtained. Those Pareto optimal solutions correspond to different classification rules, among which the user would choose the one yielding the most appropriate compromise between the cost and the expected misclassification rate.",
    keywords = "Multi-group classification, Pareto optimality, Biobjective Mixed Integer Programming, Feature cost, Support Vector Machines",
    author = "Emilio Carrizosa and Bel{\'e}n Mart{\'i}n-Barrag{\'a}n and Morales, {Dolores Romero}",
    year = "2008",
    doi = "10.1016/j.dam.2007.05.060",
    language = "English",
    volume = "156",
    pages = "950–966",
    journal = "Discrete Applied Mathematics",
    issn = "0166-218X",
    publisher = "Elsevier BV North-Holland",
    number = "6",

    }

    Multi-group Support Vector Machines with Measurement Costs : A Biobjective Approach. / Carrizosa, Emilio; Martín-Barragán, Belén; Morales, Dolores Romero.

    In: Discrete Applied Mathematics, Vol. 156, No. 6, 2008, p. 950–966.

    Research output: Contribution to journalJournal articleResearchpeer-review

    TY - JOUR

    T1 - Multi-group Support Vector Machines with Measurement Costs

    T2 - Discrete Applied Mathematics

    AU - Carrizosa,Emilio

    AU - Martín-Barragán,Belén

    AU - Morales,Dolores Romero

    PY - 2008

    Y1 - 2008

    N2 - Support Vector Machine has shown to have good performance in many practical classification settings. In this paper we propose, for multi-group classification, a biobjective optimization model in which we consider not only the generalization ability (modeled through the margin maximization), but also costs associated with the features. This cost is not limited to an economical payment, but can also refer to risk, computational effort, space requirements, etc. We introduce a Biobjective Mixed Integer Problem, for which Pareto optimal solutions are obtained. Those Pareto optimal solutions correspond to different classification rules, among which the user would choose the one yielding the most appropriate compromise between the cost and the expected misclassification rate.

    AB - Support Vector Machine has shown to have good performance in many practical classification settings. In this paper we propose, for multi-group classification, a biobjective optimization model in which we consider not only the generalization ability (modeled through the margin maximization), but also costs associated with the features. This cost is not limited to an economical payment, but can also refer to risk, computational effort, space requirements, etc. We introduce a Biobjective Mixed Integer Problem, for which Pareto optimal solutions are obtained. Those Pareto optimal solutions correspond to different classification rules, among which the user would choose the one yielding the most appropriate compromise between the cost and the expected misclassification rate.

    KW - Multi-group classification

    KW - Pareto optimality

    KW - Biobjective Mixed Integer Programming

    KW - Feature cost

    KW - Support Vector Machines

    U2 - 10.1016/j.dam.2007.05.060

    DO - 10.1016/j.dam.2007.05.060

    M3 - Journal article

    VL - 156

    SP - 950

    EP - 966

    JO - Discrete Applied Mathematics

    JF - Discrete Applied Mathematics

    SN - 0166-218X

    IS - 6

    ER -