Improving Network Intrusion Detection Using Supervised Learning for Feature Selection

Abstract

Detecting network intrusions is a crucial area of study in computer security research, with the goal of identifying malicious activities in computer networks. Prior research has concentrated on utilizing machine learning methods to spot network intrusions, typically by training models on distinct attack categories. In our study, we suggest a revised method that merges three attack categories into a unified category through supervised learning. This approach seeks to streamline the classification procedure and enhance the precision of network intrusion detection.