Analysis of Vulnerability Characteristics for Automated Penetration Testing

A report on the analysis of statistical data on vulnerabilities from public data sources has been presented to develop a mathematical model for penetration testing. Current trends in attacks and threats have been analyzed. Attack automation issues and trends in the search for an improved algorithm for automated penetration testing have been explored. A technique for assessing information security risk, considering additional criteria implemented using machine learning elements, has been proposed. A criticality criterion has been highlighted, which improves the exploit attack module in the penetration testing. Sorting identified vulnerabilities according to the proposed criteria allows improvement of the characteristics of automated penetration tests based on exploit kit attacks.