We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Enhancing the Performance of Detect DRDoS DNS Attacks Based on the Machine Learning and Proactive Feature Selection (PFS) Model.
- Authors
Nuiaa, Riyadh Rahef; Manickam, Selvakumar; Alsaeedi, Ali Hakem; Alomari, Esraa Saleh
- Abstract
The Domain Name System (DNS), which converts domain names to IP addresses, is a critical component of the internet infrastructure. Attackers exploit the existing potential vulnerabilities in this network protocol to launch their attacks. Distributed Reflection Denial of Service (DRDoS) DN S attacks are a type of Distributed Denial of Service (DDoS) attack that uses DNS vulnerabilities to carry out their attacks. These types can rapidly deplete the resources of the targeted victim system (computational and bandwidth). With the advancement of attack methods, both quantitatively and qualitatively, traditional methods used to detect DNS threats, particularly DRDoS attacks, became ineffective. Therefore, in this paper, a new model called proactive feature selection model PFS for early detection of DRDoS attacks based on DNS responses. The PFS model is divided into two stages: features selection stage and detection DRDoS attacks stage. The PFS model was validated using the standard CICDDoS2019 dataset. The results show that the PFS model achieves a high accuracy of 91.4368% and a very low FPR while reducing the number of features from 88 to 19 in the standard CICDDoS2019 dataset.
- Subjects
INTERNET domain naming system; FEATURE selection; MACHINE learning; COMPUTER network protocols; INTERNET protocol address
- Publication
IAENG International Journal of Computer Science, 2022, Vol 49, Issue 2, p511
- ISSN
1819-656X
- Publication type
Article