Machine Learning Techniques for Detecting Distributed Denial of Service (DDoS) Attacks
Cesar Marte
Co-Presenters: Individual Presentation
College: The Dorothy and George Hennings College of Science, Mathematics and Technology
Major: Information Technology
Faculty Research Mentor: Daehan Kwak
Abstract:
With the increased reliance on technology, cyber-attacks are more prevalent than ever, posing significant threats to individuals, businesses, and governments. Distributed Denial of Service (DDoS) attacks remain one of the most severe threats despite advances in security protection techniques. The objective of a DDoS attack is to overwhelm a website, server, or network resource with malicious traffic, causing service disruptions, financial losses, and reputational damage. This research focuses on the detection and classification of DDoS attacks using machine learning algorithms, a critical area of study in cybersecurity.Utilizing the CIC-IDS2017 dataset, a comprehensive collection of benign and DDoS attack data, this research aims to test and compare several traditional machine learning algorithms: K Nearest Neighbor (KNN), Support Vector Machine (SVM), Logistic Regression, Random Forest, and Gaussian Naive Bayes. These models are assessed based on their accuracy, precision, recall, and F1-score.Results indicate that most of these models can effectively distinguish between normal and attack traffic, demonstrating the potential of machine learning in enhancing cybersecurity measures. However, Gaussian Naive Bayes shows lower accuracy than others. Despite showing promising results, the practical implementation of these models faces challenges such as the need for high computational resources, susceptibility to false positives, and continuous maintenance due to emerging attack patterns.This research contributes to the ongoing efforts to improve DDoS attack detection and mitigation strategies. Future work may explore deep learning approaches and real-time implementation challenges to further advance the field of DDoS attack detection and classification.