Topic Modeling And Classification Of Common Vulnerabilities And Exposures Database

The Common Vulnerabilities and Exposures (CVEs) is a repository of publicly known cybersecurity vulnerabilities. This repository can be used in many ways including vulnerability assessment, intrusion detection, security information management, etc. Mining this repository also help us understand the trend of vulnerabilities and exposures. We have analyzed CVE entries from 2009–2019 in terms of vulnerability types defined by OWASP Top-10 risks. We extracted 121,716 unique CVEs through the twenty years. In this paper we have implemented a method to categorize these CVE entries to OWASP Top-10 risks using LDA Topic Modelling and Keyword matching technique. This method enables automatic analysis of CVE with large number of entries and contributes to the creation of vulnerability taxonomy from CVE.