Multiclass probabilistic classification for support vector machines

Ji Sang Bae; Jong Ok Kim

doi:10.1587/transinf.2014EDL8167

Multiclass probabilistic classification for support vector machines

Ji Sang Bae, Jong Ok Kim

School of Electrical Engineering

Research output: Contribution to journal › Article › peer-review

2 Citations (Scopus)

Abstract

Support Vector Machine (SVM) is one of the most widely used classifiers to categorize observations. This classifier deterministically selects a class that has the largest score for a classification output. In this letter, we propose a multiclass probabilistic classification method that reflects the degree of confidence. We apply the proposed method to age group classification and verify the performance.

Original language	English
Pages (from-to)	1251-1255
Number of pages	5
Journal	IEICE Transactions on Information and Systems
Volume	E98D
Issue number	6
DOIs	https://doi.org/10.1587/transinf.2014EDL8167
Publication status	Published - 2015 Jun 1

Bibliographical note

Publisher Copyright:
Copyright © 2015 The Institute of Electronics, Information and Communication Engineers.

Keywords

Age-group classification
Multiclass classification
Probabilistic classification
SVM

ASJC Scopus subject areas

Software
Hardware and Architecture
Computer Vision and Pattern Recognition
Electrical and Electronic Engineering
Artificial Intelligence

Access to Document

10.1587/transinf.2014EDL8167

Cite this

@article{7c174780d7f142c290110e9e8989a138,

title = "Multiclass probabilistic classification for support vector machines",

abstract = "Support Vector Machine (SVM) is one of the most widely used classifiers to categorize observations. This classifier deterministically selects a class that has the largest score for a classification output. In this letter, we propose a multiclass probabilistic classification method that reflects the degree of confidence. We apply the proposed method to age group classification and verify the performance.",

keywords = "Age-group classification, Multiclass classification, Probabilistic classification, SVM",

author = "Bae, \{Ji Sang\} and Kim, \{Jong Ok\}",

note = "Publisher Copyright: Copyright {\textcopyright} 2015 The Institute of Electronics, Information and Communication Engineers.",

year = "2015",

month = jun,

day = "1",

doi = "10.1587/transinf.2014EDL8167",

language = "English",

volume = "E98D",

pages = "1251--1255",

journal = "IEICE Transactions on Information and Systems",

issn = "0916-8532",

publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",

number = "6",

}

TY - JOUR

T1 - Multiclass probabilistic classification for support vector machines

AU - Bae, Ji Sang

AU - Kim, Jong Ok

PY - 2015/6/1

Y1 - 2015/6/1

N2 - Support Vector Machine (SVM) is one of the most widely used classifiers to categorize observations. This classifier deterministically selects a class that has the largest score for a classification output. In this letter, we propose a multiclass probabilistic classification method that reflects the degree of confidence. We apply the proposed method to age group classification and verify the performance.

AB - Support Vector Machine (SVM) is one of the most widely used classifiers to categorize observations. This classifier deterministically selects a class that has the largest score for a classification output. In this letter, we propose a multiclass probabilistic classification method that reflects the degree of confidence. We apply the proposed method to age group classification and verify the performance.

KW - Age-group classification

KW - Multiclass classification

KW - Probabilistic classification

KW - SVM

UR - https://www.scopus.com/pages/publications/84930456690

U2 - 10.1587/transinf.2014EDL8167

DO - 10.1587/transinf.2014EDL8167

M3 - Article

AN - SCOPUS:84930456690

SN - 0916-8532

VL - E98D

SP - 1251

EP - 1255

JO - IEICE Transactions on Information and Systems

JF - IEICE Transactions on Information and Systems

IS - 6

ER -

Multiclass probabilistic classification for support vector machines

Abstract

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this