The IRMA Community
Newsletters
Research IRM
Click a keyword to search titles using our InfoSci-OnDemand powered search:
|
Integrating Adversarial Training Techniques to Enhance Cybersecurity Resilience Against Machine Learning Threats
|
|
Author(s): Firas Tarik Jasim (Northern Technical University, Al-Dour Technical Institute, Iraq), Alnoman Mundher Tayyeh (Institute of Technology, Middle Technical University, Iraq), Abdulsattar Abdullah Hamad (College of Education, University of Samarra, Iraq), Omar Azeez Abbas (College of Administration and Economics, University of Samarra, Iraq), Noor Mohammed Kadhim (College Education for Human Sciences, Wasit University, Iraq), Ahmed Ibrahim Turki (University of Samarra, Iraq), Elaf Sabah Abdulwahid (College of Education for Women, Tikrit University, Iraq), Khalid Saeed Lateef Al-Badri (College of Education, Samarra University, Iraq)and Luma Saad Abdalbaqi (College of Education for Women, Tikrit University, Iraq)
Copyright: 2025
Pages: 18
Source title:
Integrating Intelligent Control Systems With Sensor Technologies
Source Author(s)/Editor(s): Abdulsattar Abdullah Hamad (University of Samarra, Iraq), Sudan Jha (Kathmandu University, Nepal)and Khalid Al-Badri (University of Samarra, Iraq)
DOI: 10.4018/979-8-3373-0330-7.ch012
Purchase
|
Abstract
The speedy improvement of gadget mastering technology has considerably transformed the landscape of cybersecurity. However, those structures are increasingly more liable to antagonistic assaults that make the maximum their weaknesses, posing enormous dangers to their effectiveness. This look investigates the mixture of hostile training strategies into device studying models to decorate their resilience against evolving cybersecurity threats. Our findings display a remarkable decline in overall performance while models are uncovered to adverse examples, with benign detection quotes within the IDS losing from 90% to 80%. In evaluation, the phishing detection tool demonstrates an ability to evolve through retraining, with accuracy increasing from 88% to 93% after imposing non-stop learning strategies. We advise a feedback loop for non-stop gaining knowledge. The outcomes underscore the need for ongoing variation in gadget studying fashions to protect in opposition to ultra-modern cyber threats, supplying treasured insights for future studies and realistic applications in cybersecurity.
Related Content
|
Licheng Huang, Bochen Xue, Yiming Chen, Peihang Wu, Yuezhong Wang, Aquil Mirza Mohammed.
© 2026.
34 pages.
|
|
Hong Rui Zhou, Min Hao Ling, Tong Yao Li, Xiang Li, Yi Ran Wu, Cong Wu.
© 2026.
34 pages.
|
|
Chenyu Liu, Yaxin Luo, Jingyan Zeng, Liyuan Fan, Mingyuan Tang, Cong Wu.
© 2026.
28 pages.
|
|
Haochen Shi, Xuan Luo, Junhao Huang, Yixiong Feng, Zihan Meng, Aquil Mirza Mohammed.
© 2026.
34 pages.
|
|
Ruiman Huang, Shuxin Jia, Zeyu Min, Haoyue Zhang, Hewa Majeed Zangana.
© 2026.
32 pages.
|
|
Shu Kei Ling, Pak Sun Wong, Kwan Ho Yuen, Mohammad Al Khaldy.
© 2026.
40 pages.
|
|
Enlong Dong, Huakun Huang, Huakai Huang, Ruize Liu, Hengxian Li.
© 2026.
34 pages.
|
|
|