Enhanced Study of Deep Learning Algorithms for Web Vulnerability Scanner

Abstract

The detection of online vulnerabilities is the most important task for network security. In this paper, deep learning methodologies for dealing with tough or complicated challenges are investigated using convolutional neural networks, long-short-term memory, and generative adversarial networks.Experimental results demonstrate that deep learning approaches can significantly outperform standard methods when compared to them. In addition, we examine the various aspects that affect performance. This work can provide researchers with useful direction when designing network architecture and parameters for identifying web attacks.