Usage of Artificial Intelligence in DOS/DDOS Attack Detection

Abstract

The main objective of a Denial of Service (DOS) attacks is to target a specific entry and create a flood of different type of network packets. If the attack is formed in Distributed Denial of Service than the attacker compiles multiple systems across the internet called as zombies/agents and executes the attack by remotely controlling them. In this paper it is aimed to examine detection of Denial of Service attacks through different Machine Learning Algorithms and Artificial Neural Networks (ANN). The evaluation will be done with the Knowledge Discovery and Data Mining Tools Competition (KDD 99) dataset and the data collected in lab tests. The focus of the study will be the assessment of the Machine Learning and ANN algorithms success in the detection of Network Layer DOS Attacks.