RV-SVM: An efficient method for learning ranking SVM

Yu Hwanjo; Kim Youngdae; Hwang Seungwon

doi:10.1007/978-3-642-01307-2_39

RV-SVM: An efficient method for learning ranking SVM

Yu Hwanjo, Kim Youngdae, Hwang Seungwon

College of Computing

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

11 Citations (Scopus)

Abstract

Learning ranking (or preference) functions has become an important data mining task in recent years, as various applications have been found in information retrieval. Among rank learning methods, ranking SVMhas been favorably applied to various applications, e.g., optimizing search engines, improving data retrieval quality. In this paper, we first develop a 1-norm ranking SVM that is faster in testing than the standard ranking SVM, and propose Ranking Vector SVM (RV-SVM) that revises the 1-norm ranking SVM for faster training. The number of variables in the RV-SVM is significantly smaller, thus the RV-SVM trains much faster than the other ranking SVMs.We experimentally compared the RV-SVM with the state-of-the-art rank learning method provided in SVM-light. The RV-SVMuses much less support vectors and trains much faster for nonlinear kernels than the SVM-light. The accuracies of RV-SVM and SVM-light are comparable on relatively large data sets. Our implementation of RV-SVM is posted at http://iis.postech.ac.kr/rv-svm.

Original language	English
Title of host publication	13th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2009
Pages	426-438
Number of pages	13
DOIs	https://doi.org/10.1007/978-3-642-01307-2_39
Publication status	Published - 2009
Event	13th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2009 - Bangkok, Thailand Duration: 2009 Apr 27 → 2009 Apr 30

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	5476 LNAI
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Other

Other	13th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2009
Country/Territory	Thailand
City	Bangkok
Period	09/4/27 → 09/4/30

Bibliographical note

Funding Information:
This work was supported by the Korea Research Foundation Grant funded by the Korean Government (KRF-2008-331-D00528).

All Science Journal Classification (ASJC) codes

Theoretical Computer Science
Computer Science(all)

Access to Document

10.1007/978-3-642-01307-2_39

Cite this

Hwanjo, Y., Youngdae, K., & Seungwon, H. (2009). RV-SVM: An efficient method for learning ranking SVM. In 13th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2009 (pp. 426-438). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5476 LNAI). https://doi.org/10.1007/978-3-642-01307-2_39

@inproceedings{df1b25b07ff74d018779cd34c1cd1e93,

title = "RV-SVM: An efficient method for learning ranking SVM",

abstract = "Learning ranking (or preference) functions has become an important data mining task in recent years, as various applications have been found in information retrieval. Among rank learning methods, ranking SVMhas been favorably applied to various applications, e.g., optimizing search engines, improving data retrieval quality. In this paper, we first develop a 1-norm ranking SVM that is faster in testing than the standard ranking SVM, and propose Ranking Vector SVM (RV-SVM) that revises the 1-norm ranking SVM for faster training. The number of variables in the RV-SVM is significantly smaller, thus the RV-SVM trains much faster than the other ranking SVMs.We experimentally compared the RV-SVM with the state-of-the-art rank learning method provided in SVM-light. The RV-SVMuses much less support vectors and trains much faster for nonlinear kernels than the SVM-light. The accuracies of RV-SVM and SVM-light are comparable on relatively large data sets. Our implementation of RV-SVM is posted at http://iis.postech.ac.kr/rv-svm.",

author = "Yu Hwanjo and Kim Youngdae and Hwang Seungwon",

note = "Funding Information: This work was supported by the Korea Research Foundation Grant funded by the Korean Government (KRF-2008-331-D00528).; 13th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2009 ; Conference date: 27-04-2009 Through 30-04-2009",

year = "2009",

doi = "10.1007/978-3-642-01307-2_39",

language = "English",

isbn = "3642013066",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

pages = "426--438",

booktitle = "13th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2009",

}

Hwanjo, Y, Youngdae, K & Seungwon, H 2009, RV-SVM: An efficient method for learning ranking SVM. in 13th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2009. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 5476 LNAI, pp. 426-438, 13th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2009, Bangkok, Thailand, 09/4/27. https://doi.org/10.1007/978-3-642-01307-2_39

RV-SVM: An efficient method for learning ranking SVM. / Hwanjo, Yu; Youngdae, Kim; Seungwon, Hwang.
13th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2009. 2009. p. 426-438 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5476 LNAI).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - RV-SVM

T2 - 13th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2009

AU - Hwanjo, Yu

AU - Youngdae, Kim

AU - Seungwon, Hwang

N1 - Funding Information: This work was supported by the Korea Research Foundation Grant funded by the Korean Government (KRF-2008-331-D00528).

PY - 2009

Y1 - 2009

N2 - Learning ranking (or preference) functions has become an important data mining task in recent years, as various applications have been found in information retrieval. Among rank learning methods, ranking SVMhas been favorably applied to various applications, e.g., optimizing search engines, improving data retrieval quality. In this paper, we first develop a 1-norm ranking SVM that is faster in testing than the standard ranking SVM, and propose Ranking Vector SVM (RV-SVM) that revises the 1-norm ranking SVM for faster training. The number of variables in the RV-SVM is significantly smaller, thus the RV-SVM trains much faster than the other ranking SVMs.We experimentally compared the RV-SVM with the state-of-the-art rank learning method provided in SVM-light. The RV-SVMuses much less support vectors and trains much faster for nonlinear kernels than the SVM-light. The accuracies of RV-SVM and SVM-light are comparable on relatively large data sets. Our implementation of RV-SVM is posted at http://iis.postech.ac.kr/rv-svm.

AB - Learning ranking (or preference) functions has become an important data mining task in recent years, as various applications have been found in information retrieval. Among rank learning methods, ranking SVMhas been favorably applied to various applications, e.g., optimizing search engines, improving data retrieval quality. In this paper, we first develop a 1-norm ranking SVM that is faster in testing than the standard ranking SVM, and propose Ranking Vector SVM (RV-SVM) that revises the 1-norm ranking SVM for faster training. The number of variables in the RV-SVM is significantly smaller, thus the RV-SVM trains much faster than the other ranking SVMs.We experimentally compared the RV-SVM with the state-of-the-art rank learning method provided in SVM-light. The RV-SVMuses much less support vectors and trains much faster for nonlinear kernels than the SVM-light. The accuracies of RV-SVM and SVM-light are comparable on relatively large data sets. Our implementation of RV-SVM is posted at http://iis.postech.ac.kr/rv-svm.

UR - http://www.scopus.com/inward/record.url?scp=67650686496&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=67650686496&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-01307-2_39

DO - 10.1007/978-3-642-01307-2_39

M3 - Conference contribution

AN - SCOPUS:67650686496

SN - 3642013066

SN - 9783642013065

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 426

EP - 438

BT - 13th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2009

Y2 - 27 April 2009 through 30 April 2009

ER -

RV-SVM: An efficient method for learning ranking SVM

Abstract

Publication series

Other

Bibliographical note

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this