Enhanced Cybersecurity: Detecting ARP Spoofing Using Machine Learning Techniques

Abstract

In today’s digital world,where digital communication becomes more important, robust cybersecurity measures become more necessary than ever. Address Resolution Protocol (ARP) Spoofing, a subset of Man-in-the-Middle (MitM) attacks gives a significant threat by exploiting vulnerabilities in the Address Resolution Protocol (ARP) to intercept and manipulate data traffic. Traditional methods of detection prove inadequate against such sophisticated attacks. This paper presents a conceptual framework applying machine learning methods to enhance ARP Spoofing attack detection on end-user devices.Approach of the paper considers a range of machine learning models including, Decision Trees, Support Vector Machine (SVM), Random Forest, Neural Networks, and Long Short-Term Memory (LSTM) net works,aiming to achieve high detection accuracy and minimize false positives.The Results are promising for machine learning techniques in enhancing cybersecurity defense and mitigating ARP Spoofing-related risks