Malmö University Publications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Active Machine Learning Adversarial Attack Detection in the User Feedback Process
Malmö University, Faculty of Technology and Society (TS), Department of Computer Science and Media Technology (DVMT). Malmö University, Internet of Things and People (IOTAP).ORCID iD: 0000-0003-4071-4596
Uppsala Universitet. (Division of Scientific Computing, Department of Information Technology)
University of Tartu. (Delta Research Center, Data Systems Group)
Malmö University, Faculty of Technology and Society (TS), Department of Computer Science and Media Technology (DVMT). Malmö University, Internet of Things and People (IOTAP).ORCID iD: 0000-0002-9471-8405
2021 (English)In: IEEE Access, E-ISSN 2169-3536, E-ISSN 2169-3536, Vol. 9Article in journal (Refereed) Published
Abstract [en]

Modern Information and Communication Technology (ICT)-based applications utilize currenttechnological advancements for purposes of streaming data, as a way of adapting to the ever-changingtechnological landscape. Such efforts require providing accurate, meaningful, and trustworthy output fromthe streaming sensors particularly during dynamic virtual sensing. However, to ensure that the sensingecosystem is devoid of any sensor threats or active attacks, it is paramount to implement secure real-timestrategies. Fundamentally, real-time detection of adversarial attacks/instances during the User FeedbackProcess (UFP) is the key to forecasting potential attacks in active learning. Also, according to existingliterature, there lacks a comprehensive study that has a focus on adversarial detection from an activemachine learning perspective at the time of writing this paper. Therefore, the authors posit the importance ofdetecting adversarial attacks in active learning strategy. Attack in the context of this paper through a UFPThreat driven model has been presented as any action that exerts an alteration to the learning system ordata. To achieve this, the study employed ambient data collected from a smart environment human activityrecognition from (Continuous Ambient Sensors Dataset, CASA) with fully labeled connections, where weintentionally subject the Dataset to wrong labels as a targeted/manipulative attack (by a malevolent labeler)in the UFP, with an assumption that the user-labels were connected to unique identities. While the dataset’sfocus is to classify tasks and predict activities, our study gives a focus on active adversarial strategies froman information security point of view. Furthermore, the strategies for modeling threats have been presentedusing the Meta Attack Language (MAL) compiler for purposes adversarial detection. The findings fromthe experiments conducted have shown that real-time adversarial identification and profiling during the UFPcould significantly increase the accuracy during the learning process with a high degree of certainty and pavesthe way towards an automated adversarial detection and profiling approaches on the Internet of CognitiveThings (ICoT).

Place, publisher, year, edition, pages
IEEE, 2021. Vol. 9
Keywords [en]
Adversarial detection, user-feedback-process, active machine learning, monitoring industrial feedback.
National Category
Electrical Engineering, Electronic Engineering, Information Engineering
Research subject
Science education
Identifiers
URN: urn:nbn:se:mau:diva-41020DOI: 10.1109/ACCESS.2021.3063002ISI: 000626493900001Scopus ID: 2-s2.0-85102241032OAI: oai:DiVA.org:mau-41020DiVA, id: diva2:1534896
Available from: 2021-03-05 Created: 2021-03-05 Last updated: 2024-02-05Bibliographically approved

Open Access in DiVA

fulltext(3277 kB)387 downloads
File information
File name FULLTEXT01.pdfFile size 3277 kBChecksum SHA-512
3893092695e6a3b2ddca9cb33989b7255fa9e6b293a0c09d6aaddb9dc574447f3990c3beac888607c0e4d124cf22de58d2b44fb6e9db1a6fed83357952b49dbc
Type fulltextMimetype application/pdf

Other links

Publisher's full textScopus

Authority records

Kebande, Victor R.Persson, Jan A.

Search in DiVA

By author/editor
Kebande, Victor R.Persson, Jan A.
By organisation
Department of Computer Science and Media Technology (DVMT)Internet of Things and People (IOTAP)
In the same journal
IEEE Access
Electrical Engineering, Electronic Engineering, Information Engineering

Search outside of DiVA

GoogleGoogle Scholar
Total: 387 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

doi
urn-nbn

Altmetric score

doi
urn-nbn
Total: 57 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf