Classification and Analysis of Adversarial Machine Learning Attacks in IoT: a Label Flipping Attack Case Study
With the increased usage of Internet of Things (IoT) devices in recent years, various Machine Learning (ML) algorithms have also developed dramatically for attack detection in this domain. However, the ML models are exposed to different classes of adversarial attacks that aim to fool a model into ma...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
FRUCT
2022-11-01
|
| Series: | Proceedings of the XXth Conference of Open Innovations Association FRUCT |
| Subjects: | |
| Online Access: | https://www.fruct.org/publications/volume-32/fruct32/files/Abr.pdf |
| _version_ | 1811214883296903168 |
|---|---|
| author | Mahdi Abrishami Sajjad Dadkhah Euclides Neto Pulei Xiong Shahrear Iqbal Suprio Ray Ali Ghorbani |
| author_facet | Mahdi Abrishami Sajjad Dadkhah Euclides Neto Pulei Xiong Shahrear Iqbal Suprio Ray Ali Ghorbani |
| author_sort | Mahdi Abrishami |
| collection | DOAJ |
| description | With the increased usage of Internet of Things (IoT) devices in recent years, various Machine Learning (ML) algorithms have also developed dramatically for attack detection in this domain. However, the ML models are exposed to different classes of adversarial attacks that aim to fool a model into making an incorrect prediction. For instance, label manipulation or label flipping is an adversarial attack where the adversary attempts to manipulate the label of training data that causes the trained model biased and/or with decreased performance. However, the number of samples to be flipped in this category of attack can be restricted, giving the attacker a limited target selection. Due to the great significance of securing ML models against Adversarial Machine Learning (AML) attacks particularly in the IoT domain, this research presents an extensive review of AML in IoT. Then, a classification of AML attacks is presented based on the literature which sheds light on the future research in this domain. Next, this paper investigates the negative impact levels of applying the malicious label-flipping attacks on IoT data. We devise label-flipping scenarios for training a Support Vector Machine (SVM) model. The experiments demonstrate that the label flipping attacks impact the performance of ML models. These results can lead to designing more effective and powerful attack and defense mechanisms in adversarial settings. Finally, we show the weaknesses of the K-NN defense method against the random label flipping attack. |
| first_indexed | 2024-04-12T06:11:52Z |
| format | Article |
| id | doaj.art-9306ce0347e0494e96d8416506dd3c19 |
| institution | Directory Open Access Journal |
| issn | 2305-7254 2343-0737 |
| language | English |
| last_indexed | 2024-04-12T06:11:52Z |
| publishDate | 2022-11-01 |
| publisher | FRUCT |
| record_format | Article |
| series | Proceedings of the XXth Conference of Open Innovations Association FRUCT |
| spelling | doaj.art-9306ce0347e0494e96d8416506dd3c192022-12-22T03:44:41ZengFRUCTProceedings of the XXth Conference of Open Innovations Association FRUCT2305-72542343-07372022-11-0132131410.23919/FRUCT56874.2022.9953823Classification and Analysis of Adversarial Machine Learning Attacks in IoT: a Label Flipping Attack Case StudyMahdi Abrishami0Sajjad Dadkhah1Euclides Neto2Pulei Xiong3Shahrear Iqbal4Suprio Ray5Ali Ghorbani6Canadian Institute for Cybersecurity (CIC) / Faculty of Computer Science / University of New Brunswick (UNB), CanadaCanadian Institute for Cybersecurity (CIC) / Faculty of Computer Science / University of New Brunswick (UNB), CanadaCanadian Institute for Cybersecurity (CIC) / Faculty of Computer Science / University of New Brunswick (UNB), CanadaNational Research Council Canada, CanadaNational Research Council Canada, CanadaNational Research Council Canada, CanadaNational Research Council Canada, CanadaWith the increased usage of Internet of Things (IoT) devices in recent years, various Machine Learning (ML) algorithms have also developed dramatically for attack detection in this domain. However, the ML models are exposed to different classes of adversarial attacks that aim to fool a model into making an incorrect prediction. For instance, label manipulation or label flipping is an adversarial attack where the adversary attempts to manipulate the label of training data that causes the trained model biased and/or with decreased performance. However, the number of samples to be flipped in this category of attack can be restricted, giving the attacker a limited target selection. Due to the great significance of securing ML models against Adversarial Machine Learning (AML) attacks particularly in the IoT domain, this research presents an extensive review of AML in IoT. Then, a classification of AML attacks is presented based on the literature which sheds light on the future research in this domain. Next, this paper investigates the negative impact levels of applying the malicious label-flipping attacks on IoT data. We devise label-flipping scenarios for training a Support Vector Machine (SVM) model. The experiments demonstrate that the label flipping attacks impact the performance of ML models. These results can lead to designing more effective and powerful attack and defense mechanisms in adversarial settings. Finally, we show the weaknesses of the K-NN defense method against the random label flipping attack.https://www.fruct.org/publications/volume-32/fruct32/files/Abr.pdfiotadversarial machine learningclassification of adversarial attackslabel flippinglabel manipulation |
| spellingShingle | Mahdi Abrishami Sajjad Dadkhah Euclides Neto Pulei Xiong Shahrear Iqbal Suprio Ray Ali Ghorbani Classification and Analysis of Adversarial Machine Learning Attacks in IoT: a Label Flipping Attack Case Study Proceedings of the XXth Conference of Open Innovations Association FRUCT iot adversarial machine learning classification of adversarial attacks label flipping label manipulation |
| title | Classification and Analysis of Adversarial Machine Learning Attacks in IoT: a Label Flipping Attack Case Study |
| title_full | Classification and Analysis of Adversarial Machine Learning Attacks in IoT: a Label Flipping Attack Case Study |
| title_fullStr | Classification and Analysis of Adversarial Machine Learning Attacks in IoT: a Label Flipping Attack Case Study |
| title_full_unstemmed | Classification and Analysis of Adversarial Machine Learning Attacks in IoT: a Label Flipping Attack Case Study |
| title_short | Classification and Analysis of Adversarial Machine Learning Attacks in IoT: a Label Flipping Attack Case Study |
| title_sort | classification and analysis of adversarial machine learning attacks in iot a label flipping attack case study |
| topic | iot adversarial machine learning classification of adversarial attacks label flipping label manipulation |
| url | https://www.fruct.org/publications/volume-32/fruct32/files/Abr.pdf |
| work_keys_str_mv | AT mahdiabrishami classificationandanalysisofadversarialmachinelearningattacksiniotalabelflippingattackcasestudy AT sajjaddadkhah classificationandanalysisofadversarialmachinelearningattacksiniotalabelflippingattackcasestudy AT euclidesneto classificationandanalysisofadversarialmachinelearningattacksiniotalabelflippingattackcasestudy AT puleixiong classificationandanalysisofadversarialmachinelearningattacksiniotalabelflippingattackcasestudy AT shahreariqbal classificationandanalysisofadversarialmachinelearningattacksiniotalabelflippingattackcasestudy AT suprioray classificationandanalysisofadversarialmachinelearningattacksiniotalabelflippingattackcasestudy AT alighorbani classificationandanalysisofadversarialmachinelearningattacksiniotalabelflippingattackcasestudy |