MODEL OF CYBERATTACKS PREDICTION ON PROXY SERVERS USING A HYBRID OF RADIAL BASIS FUNCTION AND SUPPORT VECTOR MACHINE TECHNIQUES

As cyber-attacks continue to proliferate in the digital landscape, organizations face escalating threats to their networks,
particularly through proxy servers. Proxy-servers are prone to attacks such as Denial of Service (DoS) and Distributed Denial
of Service (DDoS) and existing detection and prediction systems are inefficient. Therefore, this study used a hybrid of radial
basis function (RBF) and support vector machine (SVM) techniques to predict DoS and DDoS attacks on a proxy server. The
Dynamic Systems Development Methodology was used for the design of the system. The mathematical formulation of the
hybrid model was implemented in Python and JavaScript respectively and made to run on a local area network. The result of
this research is the development of a proactive predictive security model that will helps to provide a valuable contribution to
the field of cybersecurity by combining RBF and SVM techniques for predictive analysis. The developed system has an accuracy
of 99.56% as compared to the existing individual RBF and SVM models with 77.22% and 80.00% respectively. Hence, the
developed system can effectively predict cyberattacks on proxy servers, for improved security ensuring the integrity and
availability of vital network resources.