Support vector machines with example dependent costs

Publikation: Beiträge in ZeitschriftenKonferenzaufsätze in FachzeitschriftenForschungbegutachtet

Standard

Support vector machines with example dependent costs. / Brefeld, Ulf; Geibel, Peter; Wysotzki, Fritz.
in: Lecture Notes in Computer Science, Jahrgang 2837, 01.01.2003, S. 23-34.

Publikation: Beiträge in ZeitschriftenKonferenzaufsätze in FachzeitschriftenForschungbegutachtet

Harvard

APA

Vancouver

Brefeld U, Geibel P, Wysotzki F. Support vector machines with example dependent costs. Lecture Notes in Computer Science. 2003 Jan 1;2837:23-34. doi: 10.1007/978-3-540-39857-8_5

Bibtex

@article{d62c64bc14e24a8ba0fe188cf1dd7589,
title = "Support vector machines with example dependent costs",
abstract = "Classical learning algorithms from the fields of artificial neural networks and machine learning, typically, do not take any costs into account or allow only costs depending on the classes of the examples that are used for learning. As an extension of class dependent costs, we consider costs that are example, i.e. feature and class dependent. We present a natural cost-sensitive extension of the support vector machine (SVM) and discuss its relation to the Bayes rule. We also derive an approach for including example dependent costs into an arbitrary cost-insensitive learning algorithm by sampling according to modified probability distributions.",
keywords = "Informatics, support vector machine, Cost matrix, Soft margin, Support Vector Machines (SVM), Dependent Cost, Business informatics",
author = "Ulf Brefeld and Peter Geibel and Fritz Wysotzki",
year = "2003",
month = jan,
day = "1",
doi = "10.1007/978-3-540-39857-8_5",
language = "English",
volume = "2837",
pages = "23--34",
journal = "Lecture Notes in Computer Science",
issn = "0302-9743",
publisher = "Springer Science and Business Media Deutschland",

}

RIS

TY - JOUR

T1 - Support vector machines with example dependent costs

AU - Brefeld, Ulf

AU - Geibel, Peter

AU - Wysotzki, Fritz

PY - 2003/1/1

Y1 - 2003/1/1

N2 - Classical learning algorithms from the fields of artificial neural networks and machine learning, typically, do not take any costs into account or allow only costs depending on the classes of the examples that are used for learning. As an extension of class dependent costs, we consider costs that are example, i.e. feature and class dependent. We present a natural cost-sensitive extension of the support vector machine (SVM) and discuss its relation to the Bayes rule. We also derive an approach for including example dependent costs into an arbitrary cost-insensitive learning algorithm by sampling according to modified probability distributions.

AB - Classical learning algorithms from the fields of artificial neural networks and machine learning, typically, do not take any costs into account or allow only costs depending on the classes of the examples that are used for learning. As an extension of class dependent costs, we consider costs that are example, i.e. feature and class dependent. We present a natural cost-sensitive extension of the support vector machine (SVM) and discuss its relation to the Bayes rule. We also derive an approach for including example dependent costs into an arbitrary cost-insensitive learning algorithm by sampling according to modified probability distributions.

KW - Informatics

KW - support vector machine

KW - Cost matrix

KW - Soft margin

KW - Support Vector Machines (SVM)

KW - Dependent Cost

KW - Business informatics

UR - http://www.scopus.com/inward/record.url?scp=9444295412&partnerID=8YFLogxK

U2 - 10.1007/978-3-540-39857-8_5

DO - 10.1007/978-3-540-39857-8_5

M3 - Conference article in journal

AN - SCOPUS:9444295412

VL - 2837

SP - 23

EP - 34

JO - Lecture Notes in Computer Science

JF - Lecture Notes in Computer Science

SN - 0302-9743

ER -

DOI

Zuletzt angesehen

Publikationen

  1. Logical-Rollenspiele
  2. Global Governance and the Interplay of Coordination and Contestation
  3. Multiobjective optimal control of fluid mixing
  4. Multi-view hidden markov perceptrons
  5. Exceeding Work
  6. Health and the intention to retire: exploring the moderating effects of human resources practices
  7. Subverting Autocracy
  8. Implementierung und langfristige Wirkungen des Projekts ‚Jedem Kind ein Instrument‘.
  9. Pennycress-corn double-cropping increases ground beetle diversity
  10. Model and Movement
  11. Playing by Their Rules
  12. Managing Real Utopias
  13. ephemera: theory & politics in organization
  14. Identification of environmentally biodegradable scaffolds for the benign design of quinolones and related substances
  15. The Social Case as a Business Case
  16. Identitäre Zweigeschlechtlichkeit
  17. Fast Supercapacitor Charging for Electromagnetic Converter Systems by Self Powered Boost Circuit
  18. Narrative approach to futures
  19. Modellierung und Implementierung eines Order2Cash-Prozesses in verteilten Systemen
  20. Pervasive Intelligence
  21. Mycorrhizal type and tree diversity affect foliar elemental pools and stoichiometry
  22. Mapping the European Space of Circulation
  23. Effect of TiBor on the grain refinement and hot tearing susceptibility of AZ91D magnesium alloy
  24. Results of disseminating an online screen for eating disorders across the U.S.
  25. Sustainability Reporting as a Consequence of Environmental Orientation
  26. Clusters of water governance problems and their effects on policy delivery
  27. Political Representation in the EU
  28. Artificial intelligence in higher education
  29. Carbon Management Accounting and Reporting in Practice
  30. From negative to positive sustainability performance measurement and assessment? A qualitative inquiry drawing on framing effects theory
  31. Too precise to pursue