Deakin home > Deakin University Library > Deakin Research Online > Adaptive spike detection for resilient data stream mining

Adaptive spike detection for resilient data stream mining

Phua, Clifton, Smith-Miles, Kate, Lee, Vincent and Gayler, Ross 2007, Adaptive spike detection for resilient data stream mining, in Data mining and analytics 2007 : proceedings of the sixth Australasian Data Mining Conference (AusDM2007), Gold Coast, Australia, 3-4 December, 2007, Australian Computer Society, Sydney, N.S.W., pp. 181-188.

Attached Files (Some files may be inaccessible until you login with your Deakin Research Online credentials)
Name Description MIMEType Size Downloads

Title Adaptive spike detection for resilient data stream mining
Author(s) Phua, Clifton
Smith-Miles, Kate
Lee, Vincent
Gayler, Ross
Conference name Australasian Data Mining Conference (6th : 2007 : Gold Coast, Australia)
Conference location Gold Coast, Australia
Conference dates 3-4 December 2007
Title of proceedings Data mining and analytics 2007 : proceedings of the sixth Australasian Data Mining Conference (AusDM2007), Gold Coast, Australia, 3-4 December, 2007
Editor(s) Christen, Peter
Kennedy, Paul J.
Li, Jiuyong
Kolyshkina, Inna
Williams, Graham J.
Publication date 2007
Conference series Australasian Data Mining Conference
Start page 181
End page 188
Publisher Australian Computer Society
Place of publication Sydney, N.S.W.
Keyword(s) adaptive spike detection
resilient data mining
data stream mining
class-label-free attributes ranking and selection
Summary Automated adversarial detection systems can fail when under attack by adversaries. As part of a resilient data stream mining system to reduce the possibility of such failure, adaptive spike detection is attribute ranking and selection without class-labels. The first part of adaptive spike detection requires weighing all attributes for spiky-ness to rank them. The second part involves filtering some attributes with extreme weights to choose the best ones for computing each example’s suspicion score. Within an identity crime detection domain, adaptive spike detection is validated on a few million real credit applications with adversarial activity. The results are F-measure curves on eleven experiments and relative weights discussion on the best experiment. The results reinforce adaptive spike detection’s effectiveness for class-label-free attribute ranking and selection.
ISBN 9781920682514
1920682511
Language eng
Field of Research 089999 Information and Computing Sciences not elsewhere classified
HERDC Research category E1 Full written paper - refereed
Copyright notice ©2007 Australian Computer Society, Inc.
Persistent URL http://hdl.handle.net/10536/DRO/DU:30008109

Document type: Conference Paper
Collection: School of Engineering and Information Technology
Connect to link resolver
 
Unless expressly stated otherwise, the copyright for items in Deakin Research Online is owned by the author, with all rights reserved.

Versions
Version Filter Type
Access Statistics: 434 Abstract Views, 0 File Downloads  -  Detailed Statistics
Created: Mon, 29 Sep 2008, 09:04:39 EST