Developing a learning system for cross-site scripting (XSS) vulnerability in web-based applications

Although XSS attack is one of the oldest attack type to Web applications, XSS still cause vulnerabilities in many applications, including popular applications and services, due to the lack of necessary security measures in the web application development process, and XSS is constantly in the "OWASP (Open Web Application Security Platform) Top Ten Web Application Security Risk" ranking. When the studies in the literature, from 2002 to 2019 on XSS are examined, it is seen that the researches are focused only on client-side or server-side XSS solutions. In recent years, the number of studies adopting machine learning techniques to prevent XSS attacks has increased and these techniques have been found to be more effective in detecting unknown attacks. The first part of this study is an introduction to include general information about XSS. Then a brief technical information is given on Web, XSS and security-related issues. Other section includes the literature on machine learning. In the last part of this study, a learning system that applies machine learning algorithms on sample data sets with machine learning techniques and determined feature sets is developed and its results are discussed.