High-dimensional and large-scale anomaly detection using a linear one-class SVM with deep learning

Erfani, Sarah M., Rajasegarar, Sutharshan, Karunasekera, Shanika and Leckie, Christopher 2016, High-dimensional and large-scale anomaly detection using a linear one-class SVM with deep learning, Pattern recognition, vol. 58, pp. 121-134, doi: 10.1016/j.patcog.2016.03.028.

Attached Files
Name Description MIMEType Size Downloads

Title High-dimensional and large-scale anomaly detection using a linear one-class SVM with deep learning
Author(s) Erfani, Sarah M.
Rajasegarar, SutharshanORCID iD for Rajasegarar, Sutharshan orcid.org/0000-0002-6559-6736
Karunasekera, Shanika
Leckie, Christopher
Journal name Pattern recognition
Volume number 58
Start page 121
End page 134
Total pages 14
Publisher Elsevier
Place of publication Amsterdam, The Netherlands
Publication date 2016-10
ISSN 0031-3203
Keyword(s) anomaly detection
outlier detection
high-dimensional data
deep belief net
deep learning
one-class SVM
feature extraction
Summary High-dimensional problem domains pose significant challenges for anomaly detection. The presence of irrelevant features can conceal the presence of anomalies. This problem, known as the '. curse of dimensionality', is an obstacle for many anomaly detection techniques. Building a robust anomaly detection model for use in high-dimensional spaces requires the combination of an unsupervised feature extractor and an anomaly detector. While one-class support vector machines are effective at producing decision surfaces from well-behaved feature vectors, they can be inefficient at modelling the variation in large, high-dimensional datasets. Architectures such as deep belief networks (DBNs) are a promising technique for learning robust features. We present a hybrid model where an unsupervised DBN is trained to extract generic underlying features, and a one-class SVM is trained from the features learned by the DBN. Since a linear kernel can be substituted for nonlinear ones in our hybrid model without loss of accuracy, our model is scalable and computationally efficient. The experimental results show that our proposed model yields comparable anomaly detection performance with a deep autoencoder, while reducing its training and testing time by a factor of 3 and 1000, respectively.
Language eng
DOI 10.1016/j.patcog.2016.03.028
Field of Research 080109 Pattern Recognition and Data Mining
Socio Economic Objective 970108 Expanding Knowledge in the Information and Computing Sciences
HERDC Research category C1.1 Refereed article in a scholarly journal
ERA Research output type C Journal article
Grant ID LP120100529
LE120100129
Copyright notice ©2016, Elsevier
Persistent URL http://hdl.handle.net/10536/DRO/DU:30083533

Connect to link resolver
 
Unless expressly stated otherwise, the copyright for items in DRO is owned by the author, with all rights reserved.

Versions
Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 118 times in TR Web of Science
Scopus Citation Count Cited 165 times in Scopus
Google Scholar Search Google Scholar
Access Statistics: 491 Abstract Views, 4 File Downloads  -  Detailed Statistics
Created: Wed, 18 May 2016, 11:57:10 EST

Every reasonable effort has been made to ensure that permission has been obtained for items included in DRO. If you believe that your rights have been infringed by this repository, please contact drosupport@deakin.edu.au.