Advanced AI Techniques For Cyber Hacking Breach Forecasting
Keywords:
AI, cyber hacking, malware attacks.Abstract
Examining cyber event data sets is a crucial approach for enhancing our comprehension of the evolving threat landscape. This is a somewhat novel research area, and numerous investigations are yet to be conducted. This study presents a statistical analysis of a dataset pertaining to breach incidents during a 12-year period (2005–2017) including cyber hacking operations, including malware attacks. We demonstrate that, contrary to existing literature, both the inter-arrival periods of hacking breach incidents and the sizes of breaches should be described using stochastic processes rather than distributions, as they display autocorrelations. We offer specific stochastic process models to fit the inter-arrival periods and breach sizes, respectively. We additionally demonstrate that these models can forecast the inter-arrival intervals and the breach magnitudes. To gain deeper insights into the evolution of hacker breach incidents, we do both qualitative and quantitative trend studies on the dataset. We derive a collection of cyber security insights, indicating that the frequency of cyber intrusions is indeed growing, yet the severity of their damage remains unchanged.
