Predicting Vulnerability Exploits in the Wild

Abstract

Every day numerous new vulnerabilities and exploits are reported for a wide variety of different software configurations. There is a big need to be able to quickly assess associated risks and sort out which vulnerabilities that are likely to be exploited in real-world attacks. A small percentage of all vulnerabilities account for almost all the observed attack volume. We use machine learning to make automatic predictions for unseen vulnerabilities based on previous exploit patterns.