Machine Learning-Based Intrusion Detection System (IDS) for Classifying Types of Attacks on Computer Networks

Abstract

Server security on a computer network is very important, maintaining the security of a computer network in
order to maintain information, data and maintain infrastructure so that it can work and function properly and provide
access rights to registered users, this research, aims to build an IDS (Intrusion Detection System) on the network and
Server using Raspberry Pi with SNORT which is useful for monitoring Server activity when an attempted attack occurs.
With the increasing complexity of network attacks carried out by attackers, intelligent and adaptive approaches are
needed to detect and overcome these threats. Traditional methods such as rule-based or signatures are often not effective
enough in the face of evolving attacks. The large amount of network traffic data makes it difficult to manually analyze
and detect attacks. Naive Bayes has a very important role in the classification and detection of network attacks, both
considered malicious and highly malicious, By using Naive Bayes, network security systems can become more proactive
and adaptive to attacks. This technology not only helps in detecting familiar attacks but also enables identification and
response to new or unknown attack techniques. Through proper classification, the system can provide better protection
and reduce the impact of attacks.