Unraveling the Complexity: Exploring Machine Learning Algorithms for DDoS Attack Analysis

A growing variety of Internet-based services, DDoS assaults are one of the riskiest types of cyberattacks. Attacks against distributed networks are frequently termed as DDoS assaults. These assaults take advantage of particular weakness each victim possesses, for instance, the layout of the official organization’s website. Many studies on Intrusion Detection System have been conducted and solutions with greater and to safeguard services against DDoS assaults, more precise detection has been proposed. Proposed a method that uses machine learning to identify assault symptoms in order to anticipate them. To determine the current situation of DDoS attacks, Utilizing the most recent dataset is crucial. In the suggested study, Python was used as a simulator and the GitHub repository provided the UNWS-np-15 dataset. This has been accomplished by using the classification algorithms Random Forest, Decision tree, XG Boost, Ensemble Technique, Max Voting. A complete structure for the prediction of DDoS assaults was proposed in order to permit the research. To gauge the effectiveness of the model, constructed a confusion matrix. The accuracy of the model has been increased approximately from 72% to 97% respectively.