A Flexible Vulnerability Reduction System Using Machine Learning

Abstract

The human system have an ability to adapt dynamically and protect against biological viruses is amazing. Computer security faces an ever-increasing threat and a system which can prevent any viruses coming in, is an open research problem. We propose a new model, called (RI Secure Web), whic h can be resilient and immune to web application vulnerability for injection and URL manipulation for injection methods using an agent based machine learning system The ability of human immune system to survive and maintain body from different damages and its self curing capabi lity inspires the development of a resilient and adaptive cyber security system. Such system functions proactive and defends itself against viruses as human immune system does. In this paper, an architectural view of a system for reducing application level vulnerabilities to protect cyber attacks, particularly injection method is proposed.