NETWORK INTRUSION DETECTION USING SUPERVISED MACHINE LEARNING TECHNIQUE

Abstract

In today’s digital era, where most services are delivered through the internet, protecting client and server machinesfrom malicious attacks is essential. Network Intrusion Detection Systems (IDS) play a critical role in identifying andmitigating such threats by analyzing incoming request data to detect potential attack signatures. This paper evaluatesthe performance of two supervised machine learning algorithms, Support Vector Machine (SVM) and Artificial NeuralNetworks (ANN), in detecting anomalies within network traffic. The IDS is trained using a comprehensive datasetcontaining normal and attack signatures. If an attack signature is detected, the request is dropped, and the maliciousdata is logged for future analysis. Through experimental analysis, we demonstrate that ANN outperforms SVM interms of accuracy, making it a more reliable choice for intrusion detection. This study highlights the importance ofenhancing IDS systems using advanced machine learning techniques to safeguard digital systems against emergingcyber threats.