A Detailed Survey on Recent XSS Web-Attacks Machine Learning Detection Techniques

Abstract

XSS attacks have become more prevalent in last few decades and thus more challenging to detect their existence. XSS attacks are broadly classified into two categories: server-based XSS attack and client-based XSS attack. Although a lot of research has already been done in this area, still the methods lack in precision and accuracy as per the literature survey. There are ample of methodologies being applied in the detection of XSS attacks using supervised learning, unsupervised learning, reinforcement learning, deep learning and metaheuristic algorithms. We present a survey of the recent approaches being applied by the numerous researchers in their proposed models. Following indexed journals were used for research papers’ collection in order to carry out a survey: Elsevier, Springer, IEEE explore, Hindawi, google scholar, and Web of Science. Moreover, in this paper, we introduce a classification chart of several machine learning algorithms that can be applied to the web-attack detection model.