Leveraging Data Mining Classifiers for Robust Ddos Attack Detection in Cloud Computing

One sort of infiltration in cloud computing environments is the Distributed Denial of Service (DDoS) attack, which covertly inserts malicious data packs into net traffic and has a devastating effect on end users. When it came to protecting the server from several external unauthorised traffics, typical cyber-security solutions fell short. A well-designed intrusion detection system (IDS) is a vital component of any loT design. When developing ML algorithms for intrusion detection, it is necessary to do thorough literature studies to study different machine learning approaches, neural network models, and optimisation algorithms. This will help uncover any gaps or flaws and ultimately lead to the development of more effective algorithms. We create data mining algorithms like C4.5, SVM, and KNN, and test how well they can identify DDoS attacks. Various classifier methods, including CART, Random Forest, and Naive Bayes, are used to compare the created algorithms' performance. While the SVM-based DDoS attack discovery model performed better than all other procedures, it struggled with handling massive datasets, according to the results analysis. An approach to this challenge is the development of Artificial Neural Network (ANN) based approaches for effective intrusion detection. The accuracy of the various approaches shown by KNN, SVM, C4.5 with KNN, and C4.5 with SVM varies when compared. In general, KNN shows respectable accuracy, although SVM is somewhat less accurate. C4.5 integrated with KNN enhances precision. Out of all the approaches that were investigated, the one that sticks out with an astonishing 97% accuracy is the one that is suggested: C4.5 with SVM.