Classification of Multiclass DDOS Attack Detection Using Bayesian Weighted Random Forest Optimized With Gazelle Optimization Algorithm

The increase in Distributed Denial of Service (DDoS) attacks poses a considerable threat to the security and stability of the current network, especially in Internet of Things (IoT) and cloud environments. Traditional detection methods often struggle with the inability to achieve a balance between detection accuracy and computational efficiency. In this manuscript, the Classification of Multiclass DDOS Attack Detection using Bayesian Weighted Random Forest Optimized with Gazelle Optimization Algorithm (DDOS-AD-BWRF-GOA) is proposed. First, the raw data is gathered from the CICDDoS2019 dataset. Then, input data are preprocessed utilizing Adaptive Bitonic Filtering for normalizing the values. The preprocessed data are fed to the Improved Feed Forward Long Short-Term Memory technique for selecting features that increase the model's execution time. The selected features are supplied to the Bayesian Weighted Random Forest (BWRF), which classifies the multiclass DDOS attack. In general, Bayesian Weighted Random Forest does not adopt any optimization methods to define optimal parameters to guarantee exact DDOS identification. Hence, GOA is proposed to optimize the Bayesian Weighted Random Forest classifier. The proposed method is implemented in MATLAB. The performance metrics, such as Accuracy, Precision, Recall, F1-score, Specificity, Error rate, and Computational time are evaluated. The proposed method attains 15.34%, 24.1%, and 18.9% higher accuracy and 12.4%, 18.24%, and 22.6% higher precision when analyzed with existing techniques: Hybrid deep learning method for DDOS detection and classification (HDL-DDOS-DC), Edge-HetIoT Defense against DDoS attack utilizing learning techniques (EHD-DDOS-LT), and Digital twin-enabled intelligent DDOS detection for autonomous core networks (DTI-DDOS-ACN), respectively.