Training Machine Learning Algorithms to Detect Distributed Denial of Service Attacks

Cyber-attacks are becoming more and more sophisticated, posing a serious threat to our technologically dependent society. Such an attack is the Distributed Denial of Service (DDoS) attack, which is becoming a serious threat to businesses that have integrated their technology with public networks since they enable numerous attackers to obtain data or provide services to major corporations or nations. When a company's servers are overloaded with fraudulent requests while legitimate users' requests are denied, Distributed Denial of Service (DDoS) attacks disrupt Web service availability for an arbitrary amount of time. This results in financial losses since services are rendered unavailable. This paper provides a comparative analysis of popular ML algorithms, including Logistic Regression, Random Forest, and Neural Network, in terms of their effectiveness in DDoS attack detection. Along with a comprehensive evaluation of its performance. The study incorporates numerical data analysis and relevant diagrams to offer insights into the comparative efficacy of different ML techniques for DDoS attack detection. Keywords: DDoS attacks, machine learning, random forest, Logistic Regression, Neural Network