Enhanced local support vector machine with fast cross-validation capability

Yu Ann Chen, Pau-Choo Chung

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Local SVM is a lazy learner combining k-nearest neighbor search and support vector machine classifier. We propose an improved implementation of local SVM which utilizes tree structure for efficient nearest neighbor search and a method to avoid unnecessary SVM training in areas far from decision boundary. The proposed lazy learner has great advantage on cross-validation efficiency while maintaining comparable accuracy to traditional SVM. The proposed method also enables us to conduct leave-one-out cross-validation which is previously considered too time-consuming to be practical on large dataset.

Original languageEnglish
Title of host publicationIntelligent Systems and Applications - Proceedings of the International Computer Symposium, ICS 2014
EditorsWilliam Cheng-Chung Chu, Stephen Jenn-Hwa Yang, Han-Chieh Chao
PublisherIOS Press
Pages491-500
Number of pages10
ISBN (Electronic)9781614994831
DOIs
Publication statusPublished - 2015 Jan 1
EventInternational Computer Symposium, ICS 2014 - Taichung, Taiwan
Duration: 2014 Dec 122014 Dec 14

Publication series

NameFrontiers in Artificial Intelligence and Applications
Volume274
ISSN (Print)0922-6389

Other

OtherInternational Computer Symposium, ICS 2014
CountryTaiwan
CityTaichung
Period14-12-1214-12-14

Fingerprint

Support vector machines
Classifiers
Nearest neighbor search

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence

Cite this

Chen, Y. A., & Chung, P-C. (2015). Enhanced local support vector machine with fast cross-validation capability. In W. C-C. Chu, S. J-H. Yang, & H-C. Chao (Eds.), Intelligent Systems and Applications - Proceedings of the International Computer Symposium, ICS 2014 (pp. 491-500). (Frontiers in Artificial Intelligence and Applications; Vol. 274). IOS Press. https://doi.org/10.3233/978-1-61499-484-8-491
Chen, Yu Ann ; Chung, Pau-Choo. / Enhanced local support vector machine with fast cross-validation capability. Intelligent Systems and Applications - Proceedings of the International Computer Symposium, ICS 2014. editor / William Cheng-Chung Chu ; Stephen Jenn-Hwa Yang ; Han-Chieh Chao. IOS Press, 2015. pp. 491-500 (Frontiers in Artificial Intelligence and Applications).
@inproceedings{c93e75b661454609936e6d6b72e92e72,
title = "Enhanced local support vector machine with fast cross-validation capability",
abstract = "Local SVM is a lazy learner combining k-nearest neighbor search and support vector machine classifier. We propose an improved implementation of local SVM which utilizes tree structure for efficient nearest neighbor search and a method to avoid unnecessary SVM training in areas far from decision boundary. The proposed lazy learner has great advantage on cross-validation efficiency while maintaining comparable accuracy to traditional SVM. The proposed method also enables us to conduct leave-one-out cross-validation which is previously considered too time-consuming to be practical on large dataset.",
author = "Chen, {Yu Ann} and Pau-Choo Chung",
year = "2015",
month = "1",
day = "1",
doi = "10.3233/978-1-61499-484-8-491",
language = "English",
series = "Frontiers in Artificial Intelligence and Applications",
publisher = "IOS Press",
pages = "491--500",
editor = "Chu, {William Cheng-Chung} and Yang, {Stephen Jenn-Hwa} and Han-Chieh Chao",
booktitle = "Intelligent Systems and Applications - Proceedings of the International Computer Symposium, ICS 2014",
address = "Netherlands",

}

Chen, YA & Chung, P-C 2015, Enhanced local support vector machine with fast cross-validation capability. in WC-C Chu, SJ-H Yang & H-C Chao (eds), Intelligent Systems and Applications - Proceedings of the International Computer Symposium, ICS 2014. Frontiers in Artificial Intelligence and Applications, vol. 274, IOS Press, pp. 491-500, International Computer Symposium, ICS 2014, Taichung, Taiwan, 14-12-12. https://doi.org/10.3233/978-1-61499-484-8-491

Enhanced local support vector machine with fast cross-validation capability. / Chen, Yu Ann; Chung, Pau-Choo.

Intelligent Systems and Applications - Proceedings of the International Computer Symposium, ICS 2014. ed. / William Cheng-Chung Chu; Stephen Jenn-Hwa Yang; Han-Chieh Chao. IOS Press, 2015. p. 491-500 (Frontiers in Artificial Intelligence and Applications; Vol. 274).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Enhanced local support vector machine with fast cross-validation capability

AU - Chen, Yu Ann

AU - Chung, Pau-Choo

PY - 2015/1/1

Y1 - 2015/1/1

N2 - Local SVM is a lazy learner combining k-nearest neighbor search and support vector machine classifier. We propose an improved implementation of local SVM which utilizes tree structure for efficient nearest neighbor search and a method to avoid unnecessary SVM training in areas far from decision boundary. The proposed lazy learner has great advantage on cross-validation efficiency while maintaining comparable accuracy to traditional SVM. The proposed method also enables us to conduct leave-one-out cross-validation which is previously considered too time-consuming to be practical on large dataset.

AB - Local SVM is a lazy learner combining k-nearest neighbor search and support vector machine classifier. We propose an improved implementation of local SVM which utilizes tree structure for efficient nearest neighbor search and a method to avoid unnecessary SVM training in areas far from decision boundary. The proposed lazy learner has great advantage on cross-validation efficiency while maintaining comparable accuracy to traditional SVM. The proposed method also enables us to conduct leave-one-out cross-validation which is previously considered too time-consuming to be practical on large dataset.

UR - http://www.scopus.com/inward/record.url?scp=84926454781&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84926454781&partnerID=8YFLogxK

U2 - 10.3233/978-1-61499-484-8-491

DO - 10.3233/978-1-61499-484-8-491

M3 - Conference contribution

T3 - Frontiers in Artificial Intelligence and Applications

SP - 491

EP - 500

BT - Intelligent Systems and Applications - Proceedings of the International Computer Symposium, ICS 2014

A2 - Chu, William Cheng-Chung

A2 - Yang, Stephen Jenn-Hwa

A2 - Chao, Han-Chieh

PB - IOS Press

ER -

Chen YA, Chung P-C. Enhanced local support vector machine with fast cross-validation capability. In Chu WC-C, Yang SJ-H, Chao H-C, editors, Intelligent Systems and Applications - Proceedings of the International Computer Symposium, ICS 2014. IOS Press. 2015. p. 491-500. (Frontiers in Artificial Intelligence and Applications). https://doi.org/10.3233/978-1-61499-484-8-491