Exploitability analysis using predictive cybersecurity framework

摘要: Managing Security is a complex process and existing research in the field of cybersecurity metrics provide limited insight into understanding impact attacks have on overall security goals an enterprise. We need new generation that can enable enterprises to react even faster order properly protect mission-critical systems midst both undiscovered disclosed vulnerabilities. In this paper, we propose practical predictive model for exploitability analysis networking environment using stochastic modeling. Our built upon trusted CVSS Exploitability framework analyze how atomic attributes namely Access Complexity, Vector Authentication make up score evolve over specific time period. formally define nonhomogeneous Markov which incorporates dependent covariates, vulnerability age discovery rate. The daily transition-probability matrices our study are estimated combination Frei's & Alhazmi Malaiya's Logistic model. An conducted show feasibility effectiveness proposed approach. approach enables apply analytics cyber improve decision making reduce risk.