Enhancing Cybersecurity: Machine Learning Approaches to DDoS Attack Classification on Modern Communication System with LIME Explanations

Distributed Denial of Service (DDoS) attacks are a prevalent threat to network security., posing significant risks by overwhelming communication systems and disrupting services. Addressing this issue demands precise and interpretable classification methods to detect and mitigate attacks effectively. In this study., a Random Forest classifier was trained on a labeled dataset of benign and DDoS network traffic to achieve high detection accuracy while integrating Local Interpretable Model-agnostic Explanations (LIME) to enhance interpretability. The model demonstrated remarkable performance metrics., achieving an accuracy of 99.95%., precision of 1.0000., recall of 0.9990., and an Fl score of 0.9995., thus indicating strong classification ability with minimal errors. Through LIME., the study also highlighted the most influential features driving classification decisions., offering actionable insights into model predictions. This combination of high accuracy and explainability aids in developing DDoS detection models that are both effective and transparent., providing valuable tools for cybersecurity efforts.