Measuring GPU-accelerated parallel SVM performance using large datasets for multi-class machine learning problem

Muhamad Abdul Hay Bin Sulaiman, Azizah Suliman, Abd Rahim Ahmad

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

1 Citation (Scopus)

Abstract

This paper presents performance evaluation of GPU-accelerated Support Vector Machines (SVMs) using large datasets. Although SVMs algorithm is popular among machine learning researchers and data mining practitioners, its computational time is too long and impractical for large datasets due to its complex Quadratic Programming (QP) solver. The result shows that using GPU-accelerated SVMs can significantly reduce computational time for training phase of SVMs and it can be a viable solution for any project that require real-time forecasting output.

Original languageEnglish
Title of host publicationConference Proceedings - 6th International Conference on Information Technology and Multimedia at UNITEN
Subtitle of host publicationCultivating Creativity and Enabling Technology Through the Internet of Things, ICIMU 2014
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages299-302
Number of pages4
ISBN (Electronic)9781479954230
DOIs
Publication statusPublished - 23 Mar 2015
Event6th International Conference on Information Technology and Multimedia, ICIMU 2014 - Putrajaya, Malaysia
Duration: 18 Nov 201420 Nov 2014

Publication series

NameConference Proceedings - 6th International Conference on Information Technology and Multimedia at UNITEN: Cultivating Creativity and Enabling Technology Through the Internet of Things, ICIMU 2014

Other

Other6th International Conference on Information Technology and Multimedia, ICIMU 2014
CountryMalaysia
CityPutrajaya
Period18/11/1420/11/14

Fingerprint

Support vector machines
Learning systems
Quadratic programming
Data mining
Graphics processing unit

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Computer Science Applications
  • Information Systems
  • Software

Cite this

Sulaiman, M. A. H. B., Suliman, A., & Ahmad, A. R. (2015). Measuring GPU-accelerated parallel SVM performance using large datasets for multi-class machine learning problem. In Conference Proceedings - 6th International Conference on Information Technology and Multimedia at UNITEN: Cultivating Creativity and Enabling Technology Through the Internet of Things, ICIMU 2014 (pp. 299-302). [7066648] (Conference Proceedings - 6th International Conference on Information Technology and Multimedia at UNITEN: Cultivating Creativity and Enabling Technology Through the Internet of Things, ICIMU 2014). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICIMU.2014.7066648
Sulaiman, Muhamad Abdul Hay Bin ; Suliman, Azizah ; Ahmad, Abd Rahim. / Measuring GPU-accelerated parallel SVM performance using large datasets for multi-class machine learning problem. Conference Proceedings - 6th International Conference on Information Technology and Multimedia at UNITEN: Cultivating Creativity and Enabling Technology Through the Internet of Things, ICIMU 2014. Institute of Electrical and Electronics Engineers Inc., 2015. pp. 299-302 (Conference Proceedings - 6th International Conference on Information Technology and Multimedia at UNITEN: Cultivating Creativity and Enabling Technology Through the Internet of Things, ICIMU 2014).
@inproceedings{41e634b85186416cb3ea3207c77c0b0a,
title = "Measuring GPU-accelerated parallel SVM performance using large datasets for multi-class machine learning problem",
abstract = "This paper presents performance evaluation of GPU-accelerated Support Vector Machines (SVMs) using large datasets. Although SVMs algorithm is popular among machine learning researchers and data mining practitioners, its computational time is too long and impractical for large datasets due to its complex Quadratic Programming (QP) solver. The result shows that using GPU-accelerated SVMs can significantly reduce computational time for training phase of SVMs and it can be a viable solution for any project that require real-time forecasting output.",
author = "Sulaiman, {Muhamad Abdul Hay Bin} and Azizah Suliman and Ahmad, {Abd Rahim}",
year = "2015",
month = "3",
day = "23",
doi = "10.1109/ICIMU.2014.7066648",
language = "English",
series = "Conference Proceedings - 6th International Conference on Information Technology and Multimedia at UNITEN: Cultivating Creativity and Enabling Technology Through the Internet of Things, ICIMU 2014",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "299--302",
booktitle = "Conference Proceedings - 6th International Conference on Information Technology and Multimedia at UNITEN",
address = "United States",

}

Sulaiman, MAHB, Suliman, A & Ahmad, AR 2015, Measuring GPU-accelerated parallel SVM performance using large datasets for multi-class machine learning problem. in Conference Proceedings - 6th International Conference on Information Technology and Multimedia at UNITEN: Cultivating Creativity and Enabling Technology Through the Internet of Things, ICIMU 2014., 7066648, Conference Proceedings - 6th International Conference on Information Technology and Multimedia at UNITEN: Cultivating Creativity and Enabling Technology Through the Internet of Things, ICIMU 2014, Institute of Electrical and Electronics Engineers Inc., pp. 299-302, 6th International Conference on Information Technology and Multimedia, ICIMU 2014, Putrajaya, Malaysia, 18/11/14. https://doi.org/10.1109/ICIMU.2014.7066648

Measuring GPU-accelerated parallel SVM performance using large datasets for multi-class machine learning problem. / Sulaiman, Muhamad Abdul Hay Bin; Suliman, Azizah; Ahmad, Abd Rahim.

Conference Proceedings - 6th International Conference on Information Technology and Multimedia at UNITEN: Cultivating Creativity and Enabling Technology Through the Internet of Things, ICIMU 2014. Institute of Electrical and Electronics Engineers Inc., 2015. p. 299-302 7066648 (Conference Proceedings - 6th International Conference on Information Technology and Multimedia at UNITEN: Cultivating Creativity and Enabling Technology Through the Internet of Things, ICIMU 2014).

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

TY - GEN

T1 - Measuring GPU-accelerated parallel SVM performance using large datasets for multi-class machine learning problem

AU - Sulaiman, Muhamad Abdul Hay Bin

AU - Suliman, Azizah

AU - Ahmad, Abd Rahim

PY - 2015/3/23

Y1 - 2015/3/23

N2 - This paper presents performance evaluation of GPU-accelerated Support Vector Machines (SVMs) using large datasets. Although SVMs algorithm is popular among machine learning researchers and data mining practitioners, its computational time is too long and impractical for large datasets due to its complex Quadratic Programming (QP) solver. The result shows that using GPU-accelerated SVMs can significantly reduce computational time for training phase of SVMs and it can be a viable solution for any project that require real-time forecasting output.

AB - This paper presents performance evaluation of GPU-accelerated Support Vector Machines (SVMs) using large datasets. Although SVMs algorithm is popular among machine learning researchers and data mining practitioners, its computational time is too long and impractical for large datasets due to its complex Quadratic Programming (QP) solver. The result shows that using GPU-accelerated SVMs can significantly reduce computational time for training phase of SVMs and it can be a viable solution for any project that require real-time forecasting output.

UR - http://www.scopus.com/inward/record.url?scp=84937434924&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84937434924&partnerID=8YFLogxK

U2 - 10.1109/ICIMU.2014.7066648

DO - 10.1109/ICIMU.2014.7066648

M3 - Conference contribution

T3 - Conference Proceedings - 6th International Conference on Information Technology and Multimedia at UNITEN: Cultivating Creativity and Enabling Technology Through the Internet of Things, ICIMU 2014

SP - 299

EP - 302

BT - Conference Proceedings - 6th International Conference on Information Technology and Multimedia at UNITEN

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Sulaiman MAHB, Suliman A, Ahmad AR. Measuring GPU-accelerated parallel SVM performance using large datasets for multi-class machine learning problem. In Conference Proceedings - 6th International Conference on Information Technology and Multimedia at UNITEN: Cultivating Creativity and Enabling Technology Through the Internet of Things, ICIMU 2014. Institute of Electrical and Electronics Engineers Inc. 2015. p. 299-302. 7066648. (Conference Proceedings - 6th International Conference on Information Technology and Multimedia at UNITEN: Cultivating Creativity and Enabling Technology Through the Internet of Things, ICIMU 2014). https://doi.org/10.1109/ICIMU.2014.7066648