Study of the SNORT intrusion detection system based on machine learning

Internet users can carry out their attacks by taking advantage of the flaws in networks and systems, therefore they are not always acting with good intentions. The repercussions of these attacks can be severe for both private individuals (information loss or worse, theft, invasion of privacy…) and businesses (know-how loss, harm to reputation, financial loss…). To do this, administrators use efficient security tools that can safeguard the company’s network. In this context, Intrusion Detection Systems (IDS) are a good alternative to better protect the computer network. An IDS is a device that secretly monitors network traffic to spot unusual or suspicious activity and take preventive measures against intrusion risks. In this paper, we will focus on network intrusion detection tools especially Snort, which can detect network intrusions in real time. The major goal of this paper is to examine the flaws in this technology and compare several works that make use of machine learning techniques frequently to address these flaws.