Learning linear classifiers sensitive to example dependent and noisy costs

Publikation: Beiträge in ZeitschriftenZeitschriftenaufsätzeForschungbegutachtet

Standard

Learning linear classifiers sensitive to example dependent and noisy costs. / Geibel, Peter; Brefeld, Ulf; Wysotzki, Fritz.
in: Lecture Notes in Computer Science, Jahrgang 2810, 01.01.2003, S. 167-178.

Publikation: Beiträge in ZeitschriftenZeitschriftenaufsätzeForschungbegutachtet

Harvard

APA

Vancouver

Geibel P, Brefeld U, Wysotzki F. Learning linear classifiers sensitive to example dependent and noisy costs. Lecture Notes in Computer Science. 2003 Jan 1;2810:167-178. doi: 10.1007/978-3-540-45231-7_16

Bibtex

@article{115755a0aa344938855089dff5b5dded,
title = "Learning linear classifiers sensitive to example dependent and noisy costs",
abstract = "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 derive a cost-sensitive perceptron learning rule for non-separable classes, that can be extended to multi-modal classes (DIPOL) and present a natural cost-sensitive extension of the support vector machine (SVM).",
keywords = "Business informatics, Informatics",
author = "Peter Geibel and Ulf Brefeld and Fritz Wysotzki",
year = "2003",
month = jan,
day = "1",
doi = "10.1007/978-3-540-45231-7_16",
language = "English",
volume = "2810",
pages = "167--178",
journal = "Lecture Notes in Computer Science",
issn = "0302-9743",
publisher = "Springer",

}

RIS

TY - JOUR

T1 - Learning linear classifiers sensitive to example dependent and noisy costs

AU - Geibel, Peter

AU - Brefeld, Ulf

AU - Wysotzki, Fritz

PY - 2003/1/1

Y1 - 2003/1/1

N2 - 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 derive a cost-sensitive perceptron learning rule for non-separable classes, that can be extended to multi-modal classes (DIPOL) and present a natural cost-sensitive extension of the support vector machine (SVM).

AB - 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 derive a cost-sensitive perceptron learning rule for non-separable classes, that can be extended to multi-modal classes (DIPOL) and present a natural cost-sensitive extension of the support vector machine (SVM).

KW - Business informatics

KW - Informatics

UR - http://www.scopus.com/inward/record.url?scp=35248866377&partnerID=8YFLogxK

UR - https://www.mendeley.com/catalogue/f68182df-2f13-3814-b398-89977fc2b1ce/

U2 - 10.1007/978-3-540-45231-7_16

DO - 10.1007/978-3-540-45231-7_16

M3 - Journal articles

AN - SCOPUS:35248866377

VL - 2810

SP - 167

EP - 178

JO - Lecture Notes in Computer Science

JF - Lecture Notes in Computer Science

SN - 0302-9743

ER -

DOI