Comparing ML Algorithms: A Comprehensive Evaluation of Intrusion, Anomaly, and DoS Attack Detection Systems

Cybersecurity serves an essential function in preserving sensitive data against continually evolving threats. The present study intends to gauge how effectively machine learning (ML) algorithms work in the three primary domains of cyber defense: intrusion detection, anomaly detection, and denial-ofservice (DoS) attack detection. A systematic assessment was conducted to assess multiple machine learning techniques using metrics for success like accuracy, precision, recall, and F1-score. These outcomes highlight the usefulness of Random Forest and Decision Tree classifiers in recognizing cyber threats and suggest that Logistic Regression offers an effective option for anomaly identification. For additional cyber security measures, future research must focus on understanding deep learning models and studying a variety of datasets.