Identifying Patterns in Common Vulnerabilities and Exposures Databases with Exploratory Data Analysis

Abstract

Cybersecurity threats are targeted by unauthorized users for data stealing, manipulation, harming the reputation of another organization, and damaging the information system. A standardized record of Vulnerability is maintained in National Vulnerability Database that is accessible to the community. Each reported security flaw’s detail can be fetched through Common Vulnerability and Exposures ID. This database is recorded for more than a decade and the detailed analysis of the reported security flaws helps network administrators and security professionals to enhance the cybersecurity practices in an organization. The manual study of the vulnerability database is a challenging task and the threat alert tools provide an automatic method of fetching the vulnerability records from the central database. Threat Intelligence is an evolving field where comprehensive analysis can be generated using artificial intelligence methodologies with previous data records. Before applying any machine learning approach to a vulnerability database for threat intelligence, it is necessary to perform exploratory data analysis of the recorded security flaws. In this paper, the authors extracted the recent vulnerability data from the national vulnerability database and performed exploratory data analysis to understand the trends in reported Common Vulnerability and Exposures to frame the future directions for threat intelligence solutions. This experimental study helps in identifying the main attributes of artificial intelligence-based threat alert generation solutions.