With the continuous increase in reported Common Vulnerabilities and Exposures (CVEs), security teams are overwhelmed by vast amounts of data, which are often analyzed manually, leading to a slow and inefficient process. To address cybersecurity threats effectively, it is essential to establish connections across multiple security entity databases, including CVEs, Common Weakness Enumeration (CWEs), and Common Attack Pattern Enumeration and Classification (CAPECs). In this study, we introduce a new approach that leverages the RotatE \cite{RotatE} knowledge graph embedding model, initialized with embeddings from Ada language model developed by OpenAI \cite{embeddingada}. Additionally, we extend this approach by initializing the embeddings for the relations. \ignore{This method surpasses previous attempts and provides a valuable tool for security teams to efficiently identify and respond to cybersecurity threats.
Unlike previous works that only handled CVEs present in the training set, our approach can deal with unseen entities. Furthermore, we contribute a comprehensive dataset and our models for future benchmarking.
DEEPNESS Lab 2022 © all rights reserved
@inproceedings{CVE–CWE,
author = {Anat Bremler–Barr and Tal Shapira and Daniel Alfasi},
title = {Next–Generation Security Entity Linkage: Harnessing the Power of Knowledge Graphs and Large Language Models},
booktitle = {Proceedings of the 16th ACM International Systems and Storage Conference (SYSTOR)},
year = {2023}, }