Evaluating Machine Learning Approaches for DDoS Attack Detection Using CIC-DDoS2019
Divya Kapil, Varsha Mittal, Durgaprasad Gangodkar · 2024
With the increasing majority of DDoS attacks, adequate detection techniques are required to protect network infrastructure. DDoS attacks are a substantial hazard to network security. The CICDDoS 2019 dataset furnishes an exhaustive collection of normal and malicious network traffic, enabling researchers to design and analyze machine learning classifiers for detecting and categorizing DDoS attacks. In this research, we present a detailed study of various machine learning techniques used for the well-known dataset, aiming to enrich the accuracy and efficiency of DDoS attack detection and categorization. We also investigate the application of machine learning techniques to detect and classify DDoS attacks, leveraging the CIC-DDoS2019 dataset for experimental validation. We assess the performance of several machine learning algorithms in determining DDoS attacks, providing a comprehensive analysis of their effectiveness. We use chi-square for feature selection and Logistic Regression, ANN and Random Forest classifier and show that Random forest got the 99.91 % accuracy. We perform comparison analysis of the performance of different models and emphasizing the significance of preprocessing steps, we seek to direct future research and growth in this field.