DDoS attacks detection using Deep Learning

Abstract

The internet faces severe threats from distributed denial of service (DDoS) attacks, which significantly impact network security. DDoS attacks aim to disrupt services by flooding servers with excessive messages, rendering them inaccessible to legitimate users. Detecting and mitigating these attacks is crucial, demanding adaptive solutions. Our project addresses these challenges by leveraging deep learning for enhanced DDoS attack detection. Through advanced model training and feature engineering, we developed a hybrid Bi-LSTM-GRU model achieving remarkable accuracy of 99.81%. This underscores deep learning’s efficacy in predictive analyticsand emphasizes the importance of feature selection for optimizing model performance. Our work contributes to smarter, responsive electronic security systems, vital for combating evolving digital threats.