An Advanced DDOS Attack Detection Model with an Ensembled SVM and Baruta Selection Technique

Abstract

The paper proposed the use of an ensembled SVM model with the Boruta selection technique to improve cloud DDoS attack detection. DDoS attacks are the most common cloud security attacks, with a 16% level of use. They can render the entire system useless, with resources offline for 24 hours, multiple days, or even a week depending on the severity of the attack. In the event of successful attacks, about $ 20,000 can be lost by a company. DDoS attacks can also make the cloud environment vulnerable to hacking, due to bad hosting or shared hosting, failure to prepare against the attack, outdated codes, and other issues. This study aims to improve the performance of Support Vector Machine (SVM) to better detect Cloud DDoS attacks by eliminating key problems and improving memory efficiency, effectiveness, and high dimensional space. Several Machine learning techniques like Decision Tree, Random Forest, KNN, and SVM were used to detect DDoS attacks in a cloud environment. In terms of detection accuracy SVM is the best among the used techniques with 84.94%. A proposed ensembled SVM with the Boruta selection technique was modeled to improve the performance of DDoS attack detection techniques in the cloud. Five different models were designed using distinct machine-learning techniques and compared to the proposed model for better performance. Logistic regression, Random Forest Classification, Support Vector Machine, K-Nearest Neighbor, and Linear Discriminant Analysis. All five Classifiers were used independently and with the Bagging technique, giving different results in all aspects. From their performance found that after the boruta selection extract 51 features out of the 79 original features of the and the data that was summed up to 1048575 was reduced to 1025 for optimal performance, Random Forest Classifier and K-Nearest Neighbor was said to perform better than the proposed SVM classifier in both  Individual modeling and with Bagging Ensembled learning. A great improvement was achieved by the model performance with a detection accuracy of 95.7%, 10.8% more than the traditional SVM, an improvement the accuracy. The implementation of KNN, Random Forest, and Linear Discriminant analysis in ensembled learning shows that their performance is better than the proposed system.