Machine Learning for Detection and Mitigation of Web Vulnerabilities and Web Attacks - Survey

Sep 17, 2026

Uploaded by Miles

Detection and mitigation of critical web vulnerabilities and attacks such as cross-site scripting (XSS) and cross-site request forgery (CSRF) are essential concerns in web security as attacks evolve and become harder to recognize. This paper surveys published research that applies classical and advanced machine learning to identify and prevent XSS and CSRF. It compares approaches, highlights key takeaways, and discusses benefits and persistent downsides to guide future research directions at the intersection of web security and machine learning.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages11

Reading Time28 min

AccessFree to download

Introduction

Cross Site Scripting

arXiv :2304 . 14451v1 [ cs .CR] 27 Apr 2023 Machine Learning for Detection and Mitigation of Web Vulnerabilities and Web Attacks Mahnoor Shahid Universita¨t des Saarlandes Saarland, Saarbru¨cken, Germany [mash00001@stud.uni-saarland.de](mash00001@stud.uni-saarland.de) Abstract—Detection and mitigation of critical web vulnerabilities and attacks like cross-site scripting (XSS), and cross-site request forgery (CSRF) have been a great concern in the 􀀂eld of web security [1]. Such web attacks are evolving and becoming more challenging to detect. Several ideas from different perspectives have been put forth that can be used to improve the performance of detecting these web vulnerabilities and preventing the attacks from happening [2]. Machine learning techniques have lately been used by researchers to defend against XSS and CSRF [3], [4], [5], and given the positive 􀀂ndings, it can be concluded that it is a promising research direction. The objective of this paper is to brie􀀃y report on the research works that have been published in this direction of applying classical and advanced machine learning to identify and prevent XSS and CSRF. The purpose of providing this survey is to address different machine learning approaches that have been implemented, understand the key takeaway of every research, discuss their positive impact and the downsides that persists, so that it can help the researchers to determine the best direction to develop new approaches for their own research and to encourage researchers to focus towards the intersection between web security and machine learning. Index Terms—Web Security; Web Vulnerabilities; Machine Learning; XSS attacks; CSRF attacks 1. Introduction Web applications have become a staple of our daily lives due to their widespread use in 􀀂elds such as banking,􀀂nance, marketing, taxation, social interaction, and even medical. Many crucial operations are carried out online on the web that includes data which could be personal details like identi􀀂cation information, passwords, pin or pass codes, medical details or 􀀂nancial details like transactions, purchase details, tax 􀀂lings etc. Although, web applications do provide security and safety measures for protection, there exists some web vulnerabilities which are loopholes, bugs or voids that are exploited, by an adversary, either on client side or server side. Indeed, the number of reported web vulnerabilities [6], [7] and the frequency of web based attacks [8] have been signi􀀂cantly rising each day, necessitating the critical need for stronger security. However, web applications, are extremely complicated and challenging to analyze because of the various complex programming techniques that are used to develop them. Therefore, identifying those web vulnerabilities is not a straight forward task and is oneof the critical component of a web application’s defense. Cross-site Scripting (XSS) and cross-site request forgery (CSRF) are considered to be among the top severe risks to web security from the kinds of vulnerabilities demonstrated by web applications, as per OWASP Top Ten (Open Web Application Security Project) [1] . Many detection algorithms based on conventional techniques, have been provided in order to track down these web vulnerabilities and defend against these attacks [2] . Unfortunately, they are not suf-􀀂ciently effective enough against a threat that can take any form and is evolving as time passes by. So, another critical component of a web application’s defense is to detect mischievous and carefully crafted web attacks by determining whether or not a target script is malicious. This is where machine learning comes to the rescue. Machine learning (as well as deep learning) has been widely acknowledged and adopted in many 􀀂elds to simplify and automate any highly complicated task, improve its performance, and increase scalability. Researchers of web security community have also tried to integrate different machine lea

FAQs

What are the main web attacks addressed in this paper?

The paper focuses on cross-site scripting (XSS) and cross-site request forgery (CSRF), which are severe web security risks.

Why is stronger detection needed for web vulnerabilities and attacks?

Reported vulnerabilities and attack frequency are rising, while conventional detection methods are not sufficiently effective against evolving threats.

What is the purpose of the survey in this research?

The survey summarizes existing machine learning approaches for detecting and mitigating XSS and CSRF, clarifying key takeaways, benefits, and persistent downsides to support future research.

Sep 17, 2026

Uploaded by Miles