Abstract:
In the field of technology and cyber security, cyber-attacks on web applications are constantly increasing. One of its most promising outcomes has been the availability of powerful web applications and low-cost internet. However, threats such as Cross site scripting, DoS, and Web attacks are becoming much more of a concern due to such technological advancements. This has essentially rekindled the interest of engineers in technological security and has prompted them to conduct extensive research on the topic.
Web applications hold vital information about both the company and its clients. However, with the rise of cybersecurity attacks, they are being hacked more often, prompting administrators to look for ways to protect them from the black hat culture. The project "Machine Learning based Open-source Web Application Firewall" is about a firewall built and deployed in a live web application. The project aims to detect incoming requests in real-time using a Machine Learning detection engine and forward them to web application firewall. It is directed towards the server if it is a benign request; otherwise, it is routed to a log, where it is analyzed.
When implemented, this project can be beneficial for administrators and network security experts to gather valuable information about a hacker's operation and its footprints, which can substantially contribute to creating new rules and mechanisms to deter potential attacks.
