Analysis of the SNORT Intrusion Detection System Using Machine Learning

Abstract

Today, cyber-attacks that exploit networks and systems vulnerabilities are becoming more and more effective, reflecting the malicious intentions of certain Internet users. These attacks harm both individuals, through loss or theft of personal data and invasion of privacy, and businesses, through loss of know-how, damage to reputation and financial loss. Against this backdrop, it is essential that network operators adopt robust security measures. Intrusion Detection Systems (IDS) are emerging as promising solutions for strengthening network security. An IDS discreetly monitors network traffic for abnormal or suspicious behavior, enabling proactive accessibility measures to be taken against intrusion attempts. This article focuses on intrusion detection technologies, and more specifically on SNORT, a tool capable of identifying network intrusions in real time. We will explore the vulnerabilities associated with this technology and look at research that applies machine learning methods to overcome these shortcomings.