Fast Kernel-based method for anomaly detection

Anh Le; Trung Le; Khanh Nguyen; Van Nguyen; Thai Hoang Le; Dat Tran

doi:10.1109/IJCNN.2016.7727609

Fast Kernel-based method for anomaly detection

Anh Le, Trung Le, Khanh Nguyen, Van Nguyen, Thai Hoang Le, Dat Tran

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

2 Citations (Scopus)

Abstract

Anomaly detection (AD) involves detecting abnormality from normality and has a wide spectrum of applications in reality. Kernel-based methods for AD have been proven robust with diverse data distributions and offering good generalization ability. Stochastic gradient descent (SGD) method has recently emerged as a promising framework to devise ultra-fast learning methods. In this paper, we conjoin the advantages of Kernel-based method and SGD-based method to propose fast learning methods for anomaly detection. We validate the proposed methods on 8 benchmark datasets in UCI repository and KDD cup 1999 dataset. The experimental results show that the proposed methods offer a comparable one-class classification accuracy while simultaneously achieving a significantly computational speed-up.

Original language	English
Title of host publication	2016 International Joint Conference on Neural Networks (IJCNN)
Subtitle of host publication	San Francisco, CA, USA — February 27 - March 02, 2016
Editors	Pablo A. Estevez
Place of Publication	New York NY USA
Publisher	IEEE, Institute of Electrical and Electronics Engineers
Pages	3211-3217
Number of pages	7
ISBN (Electronic)	9781509006199, 9781509006205
ISBN (Print)	9781509006212
DOIs	https://doi.org/10.1109/IJCNN.2016.7727609
Publication status	Published - 2016
Externally published	Yes
Event	IEEE International Joint Conference on Neural Networks 2016 - Vancouver, Canada Duration: 24 Jul 2016 → 29 Jul 2016 https://ewh.ieee.org/conf/wcci/2016/ https://ieeexplore.ieee.org/xpl/conhome/7593175/proceeding (Proceedings)

Conference

Conference	IEEE International Joint Conference on Neural Networks 2016
Abbreviated title	IJCNN 2016
Country/Territory	Canada
City	Vancouver
Period	24/07/16 → 29/07/16
Internet address	https://ewh.ieee.org/conf/wcci/2016/ https://ieeexplore.ieee.org/xpl/conhome/7593175/proceeding (Proceedings)

Keywords

Anomaly detection
Kernel method
Stochastic algorithm

Access to Document

10.1109/IJCNN.2016.7727609

Cite this

Le, A., Le, T., Nguyen, K., Nguyen, V., Le, T. H., & Tran, D. (2016). Fast Kernel-based method for anomaly detection. In P. A. Estevez (Ed.), 2016 International Joint Conference on Neural Networks (IJCNN): San Francisco, CA, USA — February 27 - March 02, 2016 (pp. 3211-3217). IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/IJCNN.2016.7727609

@inproceedings{1ac00d8bcf93446ab2ead5f6c7cb1315,

title = "Fast Kernel-based method for anomaly detection",

abstract = "Anomaly detection (AD) involves detecting abnormality from normality and has a wide spectrum of applications in reality. Kernel-based methods for AD have been proven robust with diverse data distributions and offering good generalization ability. Stochastic gradient descent (SGD) method has recently emerged as a promising framework to devise ultra-fast learning methods. In this paper, we conjoin the advantages of Kernel-based method and SGD-based method to propose fast learning methods for anomaly detection. We validate the proposed methods on 8 benchmark datasets in UCI repository and KDD cup 1999 dataset. The experimental results show that the proposed methods offer a comparable one-class classification accuracy while simultaneously achieving a significantly computational speed-up.",

keywords = "Anomaly detection, Kernel method, Stochastic algorithm",

author = "Anh Le and Trung Le and Khanh Nguyen and Van Nguyen and Le, {Thai Hoang} and Dat Tran",

year = "2016",

doi = "10.1109/IJCNN.2016.7727609",

language = "English",

isbn = "9781509006212",

pages = "3211--3217",

editor = "{A. Estevez}, {Pablo }",

booktitle = "2016 International Joint Conference on Neural Networks (IJCNN)",

publisher = "IEEE, Institute of Electrical and Electronics Engineers",

address = "United States of America",

note = "IEEE International Joint Conference on Neural Networks 2016, IJCNN 2016 ; Conference date: 24-07-2016 Through 29-07-2016",

url = "https://ewh.ieee.org/conf/wcci/2016/, https://ieeexplore.ieee.org/xpl/conhome/7593175/proceeding",

}

Le, A, Le, T, Nguyen, K, Nguyen, V, Le, TH & Tran, D 2016, Fast Kernel-based method for anomaly detection. in P A. Estevez (ed.), 2016 International Joint Conference on Neural Networks (IJCNN): San Francisco, CA, USA — February 27 - March 02, 2016. IEEE, Institute of Electrical and Electronics Engineers, New York NY USA, pp. 3211-3217, IEEE International Joint Conference on Neural Networks 2016, Vancouver, Canada, 24/07/16. https://doi.org/10.1109/IJCNN.2016.7727609

Fast Kernel-based method for anomaly detection. / Le, Anh; Le, Trung; Nguyen, Khanh et al.
2016 International Joint Conference on Neural Networks (IJCNN): San Francisco, CA, USA — February 27 - March 02, 2016. ed. / Pablo A. Estevez. New York NY USA: IEEE, Institute of Electrical and Electronics Engineers, 2016. p. 3211-3217.

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

TY - GEN

T1 - Fast Kernel-based method for anomaly detection

AU - Le, Anh

AU - Le, Trung

AU - Nguyen, Khanh

AU - Nguyen, Van

AU - Le, Thai Hoang

AU - Tran, Dat

PY - 2016

Y1 - 2016

N2 - Anomaly detection (AD) involves detecting abnormality from normality and has a wide spectrum of applications in reality. Kernel-based methods for AD have been proven robust with diverse data distributions and offering good generalization ability. Stochastic gradient descent (SGD) method has recently emerged as a promising framework to devise ultra-fast learning methods. In this paper, we conjoin the advantages of Kernel-based method and SGD-based method to propose fast learning methods for anomaly detection. We validate the proposed methods on 8 benchmark datasets in UCI repository and KDD cup 1999 dataset. The experimental results show that the proposed methods offer a comparable one-class classification accuracy while simultaneously achieving a significantly computational speed-up.

AB - Anomaly detection (AD) involves detecting abnormality from normality and has a wide spectrum of applications in reality. Kernel-based methods for AD have been proven robust with diverse data distributions and offering good generalization ability. Stochastic gradient descent (SGD) method has recently emerged as a promising framework to devise ultra-fast learning methods. In this paper, we conjoin the advantages of Kernel-based method and SGD-based method to propose fast learning methods for anomaly detection. We validate the proposed methods on 8 benchmark datasets in UCI repository and KDD cup 1999 dataset. The experimental results show that the proposed methods offer a comparable one-class classification accuracy while simultaneously achieving a significantly computational speed-up.

KW - Anomaly detection

KW - Kernel method

KW - Stochastic algorithm

UR - http://www.scopus.com/inward/record.url?scp=85007235951&partnerID=8YFLogxK

U2 - 10.1109/IJCNN.2016.7727609

DO - 10.1109/IJCNN.2016.7727609

M3 - Conference Paper

AN - SCOPUS:85007235951

SN - 9781509006212

SP - 3211

EP - 3217

BT - 2016 International Joint Conference on Neural Networks (IJCNN)

A2 - A. Estevez, Pablo

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - New York NY USA

T2 - IEEE International Joint Conference on Neural Networks 2016

Y2 - 24 July 2016 through 29 July 2016

ER -

Fast Kernel-based method for anomaly detection

Abstract

Conference

Keywords

Access to Document

Other files and links

Cite this