Vulnerability and Attack Repository for IoT: Addressing Challenges and Opportunities in Internet of Things Vulnerability Databases

Abstract

The article’s primary purpose is to highlight the importance of cybersecurity for Internet of Things (IoT) devices. Due to the widespread use of such devices in everyone’s daily and professional lives, taking care of their security is essential. This security can be strengthened by raising awareness about the vulnerabilities and risks of these devices among their manufacturers and users. Therefore, this paper shows the results of several years of work regarding building vulnerabilities and exploiting databases, with a particular focus on IoT devices. We highlight multiple unique features of our solution, such as aggregation, correlation, and enrichment of known vulnerabilities and exploits collected from 12 different sources, presentation of a timeline of threats, and combining vulnerability information with exploits. Our databases have more than 300,000 entries, which are the result of aggregating and correlating more than 1,300,000 entries from 12 different databases simultaneously, enriched with information from unstructured sources. We cover the innovative utilization of Artificial Intelligence (AI) to support data enrichment, examining the usage of the Light Gradient-Boosting Machine (LGBM) model to automatically predict vulnerability severity and Mistral7B to categorize vulnerable products, which, especially in the case of IoT devices, is critical due to the diversity of nomenclature. Social media and various unstructured sources are prominent places for gathering information. Retrieving data from them is much more complex than from structured databases, but the information there is normally supplied at a faster rate. Thus, we added Mastodon monitoring to enhance our threat timelines.

Keywords

• Internet of Things
• IoT
• IoT security
• vulnerabilities
• vulnerability database
• exploits