Deep Learning-Based Detection of CSRF Vulnerabilities in Web Applications

Abstract

Cross-Site Request Forgery (CSRF) attacks pose a significant threat to web applications, potentially leading to data breaches and service disruptions by exploiting user trust and enabling malicious actors to perform unauthorized actions on behalf of users. Existing approaches for CSRF vulnerability detection mainly rely on rule-based or machine learning techniques, which often face limitations in accurately identifying complex attack patterns. This article introduces a novel deep learning-based framework designed for the detection of CSRF vulnerabilities. The framework’s core functionality lies in its ability to analyze and classify incoming HTTP requests as either security-sensitive or benign. Our framework represents a pioneering effort in leveraging deep learning techniques for CSRF vulnerability detection, outperforming the Machine Learning based existing solutions.

Publication
2023 IEEE Intl Conf on Dependable, Autonomic and Secure Computing, Intl Conf on Pervasive Intelligence and Computing, Intl Conf on Cloud and Big Data Computing, Intl Conf on Cyber Science and Technology Congress (DASC/ PiCom/CBDCom/CyberSciTech)