Validating Network Security with Predictive Analytics: A Design Guide to Bridge Stochastic Modeling into a Computationally Intelligent Dashboard

Abstract

Network posture has historically relied on traditional and reactionary methods for protection. These methods most commonly consist of network segmentation, intrusion detection systems, intrusion prevention systems, and signature-based detections. However, these traditional security platforms have proven to be an inadequate deterrent to the complex threat matrix that we currently find ourselves in. It is only through computational intelligence that we can truly identify potential intrusion areas and network abnormalities. This study presents a path forward for industry professionals on how to implement this computational approach into their network security platforms, particularly through stochastic modeling and simulation. Acknowledging the complex nature of this approach, a human-centered design methodology is also outlined on how to integrate this science into the enterprise via a predictive analytical dashboard.