EVALUATING IMPROVED DEEP LEARNING APPROACHES FOR DETECTING DDOS ATTACKS

Distributed Denial of Service (DDoS) attacks have become increasingly common and sophisticated in recent years, highlighting the need for more effective detection methods. Our work initially compares the available machine learning and deep learning (DL) algorithms for DDOS attack and evaluates several new improved DL approaches for detecting DDoS attacks using a diverse dataset that includes real-world attack traffic and benign traffic. In the assessment, several state-of-the-art DL models, such as Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) networks that have had their parameters fine-tuned, are taken into consideration. This work contributes to the development of security solutions that are more resilient and efficient by providing insights into the usefulness and limits of various DL models and machine learning models for DDoS attack detection. Important parameters including accuracy, precision, recall, and F1-score are used to evaluate the effectiveness of each model. The results reveal that certain improved DL approaches demonstrate superior performance in detecting DDoS attacks, with some models achieving near-perfect detection rates. Furthermore, the findings provide valuable insights into feature importance and model selection for DDoS attack detection in real-world scenarios