Comparing Distributed Denial of Service (DDoS) Attack Classification Using Machine Learning Techniques in IoT Environment

Abstract

In the current information where everything has started to become more interconnected than ever, almost every individual in the world who are in developed, developing, and even 3rd world countries have access to the internet. IoT devices that make use of this technology become an integral part of our society. Although the conveniences that these devices bring are plentiful and benefit our society there are security concerns that must be addressed when looking at these IoT devices as they are vulnerable to different types of attacks. One of the simplest and most widely known attacks is the Denial of Service (DoS) and Distributed Denial of Service (DDoS) attack. This type of attack aims to exhaust the devices or network resources which causes them to become unusable. The purpose of this research is to compare the performance of two different Machine Learning Algorithms which are Multi-Layer Perceptron (MLP) and Support Vector Machine (SVM) in classifying DDoS attacks in an IoT environment. The public dataset which is BoT-IoT uses real-world IoT situations that will demonstrate flood attacks which are most used for DDoS attacks on IoT devices. The dataset will go through three phases which are pre-processing, implementation of the machine learning algorithm and performance measurement. The experimental result shows that the best result when it comes to classifying DDoS attacks in an IoT environment is MLP.