An efficient IDS for slow rate HTTP/2.0 DoS Attacks using one class classification
Nemalikanti Anand, M. A. Saifulla · 2023
Networks assume significant parts in present day life, and digital protection has turned into an essential examination region. That is an Intrusion Detection System (IDS) which identifies the vulnerabilities is essential for the networks. Artificial Intelligence(AI) strategies can naturally find the fundamental distinctions between ordinary information and unusual information with high precision. This paper brings an effective Machine learning based IDS on slow rate HTTP/2.0 DoS attacks in which initially data sets are used for making the model to understand from the real attack based data sets. With those features extraction carries 15 quintessential features. Finally, these extracted features are passed to three One class classifier algorithms namely Support Vector Machine(SVM), Isolation forest(IF), and Minimum Covariant Determinant(MCD). Evaluated results states that proposed one classifier algorithms outperforms better than other algorithms over various measures (accuracy:0.99, sensitivity:0.99, specificity:0.99)