Journal of Engineering Research

As cyber attack methods continue to evolve, traditional malicious traffic detection techniques have become inadequate for the security requirements in complex environments. Deep learning-based malicious traffic detection technology significantly enhances accuracy and real-time performance through automatic data feature extraction and multi-level pattern recognition. This paper focuses on the application of deep learning models in malicious traffic identification, exploring detection strategies and defense mechanisms under various mainstream network attack scenarios. The study shows that intelligent analysis methods combining deep neural networks can not only effectively identify complex attacks but also dynamically adjust defense strategies, achieving efficient protection against malicious traffic and advancing cybersecurity defense technology towards smarter development.

Deep learning; Malicious traffic detection; Network security; Intelligent defense; Anomaly detection