Multi-group Support Vector Machines with Measurement Costs

A Biobjective Approach

Emilio Carrizosa, Belén Martín-Barragán, Dolores Romero Morales

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

Resumé

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.
OriginalsprogEngelsk
TidsskriftDiscrete Applied Mathematics
Vol/bind156
Udgave nummer6
Sider (fra-til)950–966
ISSN0166-218X
DOI
StatusUdgivet - 2008
Udgivet eksterntJa

Emneord

  • Multi-group classification
  • Pareto optimality
  • Biobjective Mixed Integer Programming
  • Feature cost
  • Support Vector Machines

Citer dette

Carrizosa, Emilio ; Martín-Barragán, Belén ; Romero Morales, Dolores . / Multi-group Support Vector Machines with Measurement Costs : A Biobjective Approach. I: Discrete Applied Mathematics. 2008 ; Bind 156, Nr. 6. s. 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 {Romero Morales}, Dolores",
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; Romero Morales, Dolores .

I: Discrete Applied Mathematics, Bind 156, Nr. 6, 2008, s. 950–966.

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

TY - JOUR

T1 - Multi-group Support Vector Machines with Measurement Costs

T2 - A Biobjective Approach

AU - Carrizosa, Emilio

AU - Martín-Barragán, Belén

AU - Romero Morales, Dolores

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 -