Application Review of Network Attack Detection Based on Deep Learning

Abstract

With the rapid development of the Internet, network attacks (NA) have become increasingly diverse and sophisticated. Traditional detection methods relying on rules systems and statistical analysis now face significant challenges, including strong dependency on feature engineering and limited generalization ability. The effectiveness of detection systems is not only limited, but the difficulty of addressing new threats is also increased. Deep learning has been recognized as a novel solution for NA detection due to its powerful nonlinear feature extraction capabilities and end-to-end learning paradigm. A concise overview of deep learning fundamentals is first provided. Subsequently, the paper proceeds with a systematic review of prevailing deep learning approaches for NA detection, followed by an analysis of their current limitations and challenges in NA. Finally, promising future research directions are proposed to address the aforementioned limitations and challenges.

Related Papers

• → Overview of deep learning in medical imaging(2017)1,052 cited
• → Deep learning ensemble 2D CNN approach towards the detection of lung cancer(2023)173 cited
• → Exploring Deep Learning for View-Based 3D Model Retrieval(2020)109 cited
• → A Comprehensive Analysis of Machine Learning Techniques in Biomedical Image Processing Using Convolutional Neural Network(2022)18 cited
• → Survey of Machine Learning Applications of Convolutional Neural Networks to Medical Image Analysis(2021)2 cited