An efficient method for learning nonlinear ranking SVM functions

Hwanjo Yu, Jinha Kim, Youngdae Kim, Seungwon Hwang, Young Ho Lee

Research output: Contribution to journalArticle

22 Citations (Scopus)

Abstract

The problem of learning ranking (or preference) functions has become important in recent years as various applications have been found in information retrieval. Among the rank learning methods, RankSVM has been favorably used in various applications, e.g., optimizing search engines and improving data retrieval quality. Fast learning methods for linear RankSVM (RankSVM with a linear kernel) have been extensively developed, whereas methods for nonlinear RankSVM (RankSVM with nonlinear kernels) are lacking. This paper proposes an efficient method for learning with nonlinear kernels, called Ranking Vector SVM (RV-SVM). RV-SVM utilizes training vectors rather than pairwise difference vectors to determine the support vectors, and is thus faster to train than conventional RankSVMs. Experimental comparisons with the state-of-the-art RankSVM implementation provided in SVM-light show that RV-SVM is substantially faster for nonlinear kernels, although our method is slower for linear kernels. RV-SVM also uses far fewer support vectors, and thus the trained models are much simpler than those built by RankSVMs while maintaining comparable accuracy. Our implementation of RV-SVM is accessible at http://dm.hwanjoyu.org/rv-svm.

Original languageEnglish
Pages (from-to)37-48
Number of pages12
JournalInformation sciences
Volume209
DOIs
Publication statusPublished - 2012 Nov 20

Fingerprint

Ranking
kernel
Support Vector
Learning
Search Engine
Information Retrieval
Pairwise
Retrieval
Search engines
Information retrieval
Kernel

All Science Journal Classification (ASJC) codes

  • Software
  • Control and Systems Engineering
  • Theoretical Computer Science
  • Computer Science Applications
  • Information Systems and Management
  • Artificial Intelligence

Cite this

Yu, Hwanjo ; Kim, Jinha ; Kim, Youngdae ; Hwang, Seungwon ; Lee, Young Ho. / An efficient method for learning nonlinear ranking SVM functions. In: Information sciences. 2012 ; Vol. 209. pp. 37-48.
@article{9c00a3ceafa54337a16854a1ba35105f,
title = "An efficient method for learning nonlinear ranking SVM functions",
abstract = "The problem of learning ranking (or preference) functions has become important in recent years as various applications have been found in information retrieval. Among the rank learning methods, RankSVM has been favorably used in various applications, e.g., optimizing search engines and improving data retrieval quality. Fast learning methods for linear RankSVM (RankSVM with a linear kernel) have been extensively developed, whereas methods for nonlinear RankSVM (RankSVM with nonlinear kernels) are lacking. This paper proposes an efficient method for learning with nonlinear kernels, called Ranking Vector SVM (RV-SVM). RV-SVM utilizes training vectors rather than pairwise difference vectors to determine the support vectors, and is thus faster to train than conventional RankSVMs. Experimental comparisons with the state-of-the-art RankSVM implementation provided in SVM-light show that RV-SVM is substantially faster for nonlinear kernels, although our method is slower for linear kernels. RV-SVM also uses far fewer support vectors, and thus the trained models are much simpler than those built by RankSVMs while maintaining comparable accuracy. Our implementation of RV-SVM is accessible at http://dm.hwanjoyu.org/rv-svm.",
author = "Hwanjo Yu and Jinha Kim and Youngdae Kim and Seungwon Hwang and Lee, {Young Ho}",
year = "2012",
month = "11",
day = "20",
doi = "10.1016/j.ins.2012.03.022",
language = "English",
volume = "209",
pages = "37--48",
journal = "Information Sciences",
issn = "0020-0255",
publisher = "Elsevier Inc.",

}

An efficient method for learning nonlinear ranking SVM functions. / Yu, Hwanjo; Kim, Jinha; Kim, Youngdae; Hwang, Seungwon; Lee, Young Ho.

In: Information sciences, Vol. 209, 20.11.2012, p. 37-48.

Research output: Contribution to journalArticle

TY - JOUR

T1 - An efficient method for learning nonlinear ranking SVM functions

AU - Yu, Hwanjo

AU - Kim, Jinha

AU - Kim, Youngdae

AU - Hwang, Seungwon

AU - Lee, Young Ho

PY - 2012/11/20

Y1 - 2012/11/20

N2 - The problem of learning ranking (or preference) functions has become important in recent years as various applications have been found in information retrieval. Among the rank learning methods, RankSVM has been favorably used in various applications, e.g., optimizing search engines and improving data retrieval quality. Fast learning methods for linear RankSVM (RankSVM with a linear kernel) have been extensively developed, whereas methods for nonlinear RankSVM (RankSVM with nonlinear kernels) are lacking. This paper proposes an efficient method for learning with nonlinear kernels, called Ranking Vector SVM (RV-SVM). RV-SVM utilizes training vectors rather than pairwise difference vectors to determine the support vectors, and is thus faster to train than conventional RankSVMs. Experimental comparisons with the state-of-the-art RankSVM implementation provided in SVM-light show that RV-SVM is substantially faster for nonlinear kernels, although our method is slower for linear kernels. RV-SVM also uses far fewer support vectors, and thus the trained models are much simpler than those built by RankSVMs while maintaining comparable accuracy. Our implementation of RV-SVM is accessible at http://dm.hwanjoyu.org/rv-svm.

AB - The problem of learning ranking (or preference) functions has become important in recent years as various applications have been found in information retrieval. Among the rank learning methods, RankSVM has been favorably used in various applications, e.g., optimizing search engines and improving data retrieval quality. Fast learning methods for linear RankSVM (RankSVM with a linear kernel) have been extensively developed, whereas methods for nonlinear RankSVM (RankSVM with nonlinear kernels) are lacking. This paper proposes an efficient method for learning with nonlinear kernels, called Ranking Vector SVM (RV-SVM). RV-SVM utilizes training vectors rather than pairwise difference vectors to determine the support vectors, and is thus faster to train than conventional RankSVMs. Experimental comparisons with the state-of-the-art RankSVM implementation provided in SVM-light show that RV-SVM is substantially faster for nonlinear kernels, although our method is slower for linear kernels. RV-SVM also uses far fewer support vectors, and thus the trained models are much simpler than those built by RankSVMs while maintaining comparable accuracy. Our implementation of RV-SVM is accessible at http://dm.hwanjoyu.org/rv-svm.

UR - http://www.scopus.com/inward/record.url?scp=84862653947&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84862653947&partnerID=8YFLogxK

U2 - 10.1016/j.ins.2012.03.022

DO - 10.1016/j.ins.2012.03.022

M3 - Article

AN - SCOPUS:84862653947

VL - 209

SP - 37

EP - 48

JO - Information Sciences

JF - Information Sciences

SN - 0020-0255

ER -