Deakin University
Browse

Adaptive spike detection for resilient data stream mining

conference contribution
posted on 2007-01-01, 00:00 authored by C Phua, K Smith-Miles, V Lee, R Gayler
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.

History

Event

Australasian Data Mining Conference (6th : 2007 : Gold Coast, Australia)

Pagination

181 - 188

Publisher

Australian Computer Society

Location

Gold Coast, Australia

Place of publication

Sydney, N.S.W.

Start date

2007-12-03

End date

2007-12-04

ISBN-13

9781920682514

ISBN-10

1920682511

Language

eng

Publication classification

E1 Full written paper - refereed

Copyright notice

2007 Australian Computer Society, Inc.

Editor/Contributor(s)

P Christen, P Kennedy, J Li, I Kolyshkina, G Williams

Title of proceedings

Data mining and analytics 2007 : proceedings of the sixth Australasian Data Mining Conference (AusDM2007), Gold Coast, Australia, 3-4 December, 2007

Usage metrics

    Research Publications

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC