CSRF attack detection and prevention using ML
CSRF attack detection and prevention using ML
Jun 10, 2024
ยท
2 min read
Overview
CSRF Attack Detection and Prevention Using ML is a project focused on enhancing web security by leveraging machine learning techniques to identify and mitigate Cross-Site Request Forgery (CSRF) attacks. This project combines advanced data analysis and machine learning methodologies to develop a robust detection system and deployable API for real-time protection against CSRF threats.
CSRF attacks pose a significant threat to web applications by tricking users into executing unwanted actions on a different site where they are authenticated. This project aims to detect such attacks proactively and prevent potential security breaches through an intelligent, machine learning-based approach.
Techniques and Implementation
- Machine Learning Models:
- Applied various machine learning techniques, including supervised learning algorithms such as Decision Trees, Random Forests, Support Vector Machines (SVM), and Neural Networks.
- Conducted extensive experiments to compare the performance of these models in terms of accuracy, precision, recall, and F1-score.
- Selected the best-performing model for real-time CSRF attack detection based on evaluation metrics and computational efficiency.
- Model Deployment:
- Developed an API to deploy the trained machine learning model, allowing integration with web applications for real-time CSRF attack detection and prevention.
- Implemented the API using a FastAPI framework, ensuring it can handle high traffic and provide quick responses to incoming web requests.
- Detection and Prevention Mechanism:
- The API monitors incoming web requests and analyzes them using the trained machine learning model.
- The project resulted into a conference publication.
Publications
- Enhancing Web Security: A Comparative Analysis of Machine Learning Models for CSRF Detection. 2024 Intelligent Methods, Systems, and Applications (IMSA) DOI
Student List
- Bassem Osama
- Mohamed Ramadan
Main Supervisors
Did you find this page helpful? Consider sharing it ๐