Fixed-size least squares support vector machines: Scala implementation for large scale classification

Mandar Chandorkar, RaghvenPhDa Mall, Oliver Lauwers, Johan A.K. Suykens, Bart De Moor

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

3 Citations (Scopus)

Abstract

We propose FS-Scala, a flexible and modular Scala based implementation of the Fixed Size Least Squares Support Vector Machine (FS-LSSVM) for large data sets. The framework consists of a set of modules for (gradient and gradient free) optimization, model representation, kernel functions and evaluation of FS-LSSVM models. A kernel based Fixed-Size Least Squares Support Vector Machine (FS-LSSVM) model is implemented in the proposed framework, while heavily leveraging the parallel computing capabilities of Apache Spark. Global optimization routines like Coupled Simulated Annealing (CSA) and Grid Search are implemented and used to tune the hyper-parameters of the FS-LSSVM model. Finally, we carry out experiments on benchmark data sets and evaluate the performance of various kernel based FS-LSSVM models.

Original languageEnglish
Title of host publicationProceedings - 2015 IEEE Symposium Series on Computational Intelligence, SSCI 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages522-528
Number of pages7
ISBN (Electronic)9781479975600
DOIs
Publication statusPublished - 1 Jan 2015
Externally publishedYes
EventIEEE Symposium Series on Computational Intelligence, SSCI 2015 - Cape Town, South Africa
Duration: 8 Dec 201510 Dec 2015

Other

OtherIEEE Symposium Series on Computational Intelligence, SSCI 2015
CountrySouth Africa
CityCape Town
Period8/12/1510/12/15

Fingerprint

Support vector machines
Global optimization
Parallel processing systems
Simulated annealing
Electric sparks
Experiments

ASJC Scopus subject areas

  • Artificial Intelligence

Cite this

Chandorkar, M., Mall, R., Lauwers, O., Suykens, J. A. K., & De Moor, B. (2015). Fixed-size least squares support vector machines: Scala implementation for large scale classification. In Proceedings - 2015 IEEE Symposium Series on Computational Intelligence, SSCI 2015 (pp. 522-528). [7376656] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/SSCI.2015.83

Fixed-size least squares support vector machines : Scala implementation for large scale classification. / Chandorkar, Mandar; Mall, RaghvenPhDa; Lauwers, Oliver; Suykens, Johan A.K.; De Moor, Bart.

Proceedings - 2015 IEEE Symposium Series on Computational Intelligence, SSCI 2015. Institute of Electrical and Electronics Engineers Inc., 2015. p. 522-528 7376656.

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

Chandorkar, M, Mall, R, Lauwers, O, Suykens, JAK & De Moor, B 2015, Fixed-size least squares support vector machines: Scala implementation for large scale classification. in Proceedings - 2015 IEEE Symposium Series on Computational Intelligence, SSCI 2015., 7376656, Institute of Electrical and Electronics Engineers Inc., pp. 522-528, IEEE Symposium Series on Computational Intelligence, SSCI 2015, Cape Town, South Africa, 8/12/15. https://doi.org/10.1109/SSCI.2015.83
Chandorkar M, Mall R, Lauwers O, Suykens JAK, De Moor B. Fixed-size least squares support vector machines: Scala implementation for large scale classification. In Proceedings - 2015 IEEE Symposium Series on Computational Intelligence, SSCI 2015. Institute of Electrical and Electronics Engineers Inc. 2015. p. 522-528. 7376656 https://doi.org/10.1109/SSCI.2015.83
Chandorkar, Mandar ; Mall, RaghvenPhDa ; Lauwers, Oliver ; Suykens, Johan A.K. ; De Moor, Bart. / Fixed-size least squares support vector machines : Scala implementation for large scale classification. Proceedings - 2015 IEEE Symposium Series on Computational Intelligence, SSCI 2015. Institute of Electrical and Electronics Engineers Inc., 2015. pp. 522-528
@inproceedings{f7f61ade005946a08d55b59246c28548,
title = "Fixed-size least squares support vector machines: Scala implementation for large scale classification",
abstract = "We propose FS-Scala, a flexible and modular Scala based implementation of the Fixed Size Least Squares Support Vector Machine (FS-LSSVM) for large data sets. The framework consists of a set of modules for (gradient and gradient free) optimization, model representation, kernel functions and evaluation of FS-LSSVM models. A kernel based Fixed-Size Least Squares Support Vector Machine (FS-LSSVM) model is implemented in the proposed framework, while heavily leveraging the parallel computing capabilities of Apache Spark. Global optimization routines like Coupled Simulated Annealing (CSA) and Grid Search are implemented and used to tune the hyper-parameters of the FS-LSSVM model. Finally, we carry out experiments on benchmark data sets and evaluate the performance of various kernel based FS-LSSVM models.",
author = "Mandar Chandorkar and RaghvenPhDa Mall and Oliver Lauwers and Suykens, {Johan A.K.} and {De Moor}, Bart",
year = "2015",
month = "1",
day = "1",
doi = "10.1109/SSCI.2015.83",
language = "English",
pages = "522--528",
booktitle = "Proceedings - 2015 IEEE Symposium Series on Computational Intelligence, SSCI 2015",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Fixed-size least squares support vector machines

T2 - Scala implementation for large scale classification

AU - Chandorkar, Mandar

AU - Mall, RaghvenPhDa

AU - Lauwers, Oliver

AU - Suykens, Johan A.K.

AU - De Moor, Bart

PY - 2015/1/1

Y1 - 2015/1/1

N2 - We propose FS-Scala, a flexible and modular Scala based implementation of the Fixed Size Least Squares Support Vector Machine (FS-LSSVM) for large data sets. The framework consists of a set of modules for (gradient and gradient free) optimization, model representation, kernel functions and evaluation of FS-LSSVM models. A kernel based Fixed-Size Least Squares Support Vector Machine (FS-LSSVM) model is implemented in the proposed framework, while heavily leveraging the parallel computing capabilities of Apache Spark. Global optimization routines like Coupled Simulated Annealing (CSA) and Grid Search are implemented and used to tune the hyper-parameters of the FS-LSSVM model. Finally, we carry out experiments on benchmark data sets and evaluate the performance of various kernel based FS-LSSVM models.

AB - We propose FS-Scala, a flexible and modular Scala based implementation of the Fixed Size Least Squares Support Vector Machine (FS-LSSVM) for large data sets. The framework consists of a set of modules for (gradient and gradient free) optimization, model representation, kernel functions and evaluation of FS-LSSVM models. A kernel based Fixed-Size Least Squares Support Vector Machine (FS-LSSVM) model is implemented in the proposed framework, while heavily leveraging the parallel computing capabilities of Apache Spark. Global optimization routines like Coupled Simulated Annealing (CSA) and Grid Search are implemented and used to tune the hyper-parameters of the FS-LSSVM model. Finally, we carry out experiments on benchmark data sets and evaluate the performance of various kernel based FS-LSSVM models.

UR - http://www.scopus.com/inward/record.url?scp=84964994888&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84964994888&partnerID=8YFLogxK

U2 - 10.1109/SSCI.2015.83

DO - 10.1109/SSCI.2015.83

M3 - Conference contribution

AN - SCOPUS:84964994888

SP - 522

EP - 528

BT - Proceedings - 2015 IEEE Symposium Series on Computational Intelligence, SSCI 2015

PB - Institute of Electrical and Electronics Engineers Inc.

ER -