Cyber Attack Detection and Alert System Using Machine Learning

Abstract

In today's digital age, cyber attacks are one of the most serious risks to individuals, businesses, and governments. Attackers use system vulnerabilities to steal data, destroy infrastructure, or disrupt services. Traditional security solutions frequently fail to detect novel or undiscovered attack patterns in real time. To address this issue, this research presents a Cyber Attack Detection and Alert System based on Machine Learning (ML). The technology examines real-time network traffic and system logs to detect unusual behavior that could suggest a cyberattack. The system uses supervised learning algorithms like Random Forest and Support Vector Machine (SVM) to classify network activity as benign or malicious. In answer to this difficulty, we have built an upgraded approach: an incremental majority voting IDS system, which employs existing tools and methodologies to improve the resilience and adaptability of intrusion detection. Our system aims to improve the accuracy and effectiveness of intrusion detection in real-time scenarios by leveraging the collective decision-making power of multiple machine learning models such as the KNN classifier, Softmax Regressor, and Adaptive Random Forest classifier. This includes reducing false alarm rates. The system was tested using the NSLKDD and CICIDS2017 datasets, and it obtained 96.43% accuracy and 100% precision for the majority of attack types. It also handles unbalanced data well, making it appropriate for realworld applications. Finally, we improved the project by including a Stacking Classifier to generate even better results.