Interpretable Deep Learning for DDoS Defense: A SHAP-based Approach in Cloud Computing

This paper proposes an explainable deep learning model that combines CNNs and LSTMs to detect and mitigate DDoS attacks in cloud computing environments. The model leverages spatial and temporal patterns in network traffic data, and integrates SHAP for interpretability. Evaluation using diverse datasets confirms its effectiveness, and a Genetic Algorithm optimizes hyperparameters. The model provides actionable insights to aid security professionals in safeguarding cloud systems.